The impact of the midterm election result is going to be felt for years, to be sure. But some of the effects will be far more swift than that, likely kicking off right after the new Congress is sworn in. We're mere hours removed from the election, with some votes still being tallied, but already some in the media are handicapping who in the administration will remain in their positions and who will not, either because an angry Trump searching for a scapegoat fires them, or through impeachment efforts. Many people think that a third Trump impeachment will be the focus. I don't. And I hope that's not the route the Democrats go, though not every Techdirt writer agrees with me. I guess I'm just not all that excited to watch J.D. Vance walk into a room while Hail to the Chief plays behind him, even temporarily.

I much prefer the government go after the many valid targets for impeachment that currently sit in Trump's cabinet. And, given that both Trump and Vance are currently blaming the Iran not-a-war for the GOP loss, Pete Hegseth seems like a likely candidate to be the first to go.

Pentagon chief Pete Hegseth faces impeachment accusations when Congress returns after the midterm elections and others, including President Donald Trump, could be investigated for alleged misconduct as Democrats return to power.

There are all kinds of reasons to impeach Hegseth. His injection of Christian nationalism into the DoD is plainly unconstitutional. He's currently running a war with Iran that is without Congressional authorization and has resulted in the cavalier bombing of such dastardly targets as Iranian schoolgirls. He is almost certainly personally responsible for the murder of human trafficking victims as part of his campaign to lob bombs at boats off the coast of South America because they might be running drugs. In quick order, the military under Hegseth also managed to double-tap strike its victims in these boats to ensure the victims couldn't seek justice.

Thomas Massie already has articles of impeachment for Hegseth written. But Hegseth is also an easy target for Trump to fire, which may happen before any impeachment efforts can really get kicked off. It's not like Trump would lose much in firing Hegseth. He wasn't qualified for the position to begin with.

And neither is the other likely target of impeachment efforts: RFK Jr. The number of justifications for impeaching Kennedy are legion. He's completely failed in his role, preferring to run podcasts and host cooking shows instead of doing his job. Measles, screwworms, pertussis are all on the rise under his failed leadership. The medical community is in full revolt over his vaccine policies. He's hiding data from the public as to the results of his disaster of a tenure at HHS.

Like Hegseth, Kennedy already has impeachment articles filed against him that are ready to go.

Rep. Haley Stevens, D-Michigan, filed impeachment accusations against Health Secretary Robert F. Kennedy Jr. in December. Stevens accused Kennedy of freezing or canceling billions of dollars in research, firing health advisers and spreading conspiracy theories.

"Robert F. Kennedy Jr. has imperiled the health and safety of the American people, eroded public confidence in the nation's public health institutions and stalled decades of scientific and medical progress," Stevens wrote in the resolution.

Kennedy's ouster from HHS is long, long overdue.

None of this should be misconstrued to suggest that Democrats should take a page from Trump's playbook and go on a revenge tour. But dismantling the corrupt, inept regime at the Secretary level makes all the sense in the world. Both Hegseth and Kennedy have been absolute disasters for the country on a variety of levels. Allowing them two more years to continue to harm the country is plainly unacceptable. And, importantly, there are Republicans that would be likely allies to impeachment efforts of both of them.

The project of American restoration has to start now. Tearing down the House of MAGA must begin, brick by brick.

Perhaps then, and only then, we can truly be great again.

A last minute attempt to throw the midterm election into chaos failed. Leading up to this point, the signs were never good in the Magic 8 Ball way. The Trump administration wants people to believe California was overrun with non-citizen voters. The facts, however, refuse to comply with the narrative.

This particular ship has been sinking for a couple of weeks now. The DOJ delivered an accusation before it even delivered a list. It claimed — absolutely without facts in evidence — that there were at least 20,000 "illegal" voters in California. And by "illegal," the administration definitely meant "brown people we'd like to expel from the country."

California pushed back. First, it questioned the claim and received no definite answer. Then the administration reluctantly admitted — as it has in other states — that its first allegation may have been overstated. But rather than pull back and reassess, the DOJ went the other way.

I assure you that I am not making this next part up: it delivered a thumb drive containing a list of the 20,000 "illegal voters," along with a password written on a Post-It note… you know, the way aging parents handle login info for their cable TV-provided email accounts.

Then the state sued the DOJ, right after the DOJ demanded California investigate each and every one of these 20,000 allegations within 48 hours of being thumb-drived. Other voters sued the DOJ as well, which set it up for the string of extremely speedy failures it recently experienced.

Things already looked bad for the DOJ when its DHS-aided raids in California resulted in the arrests of 28 "illegal" voters — this in a state that has 23 million registered voters. Not content with this failure, the DOJ then claimed California election officials just weren't doing enough and demanded they do all the heavy lifting while the DOJ pursued its minuscule number of prosecutions.

With the mid-term election just a couple of days away, the courts moved speedily to handle this bullshit. As Chris Geidner reports, the first thing the DOJ did with the clock counting down to "FLIP EVERYTHING" time is shoot itself in the face while attempting to shoot itself in the foot:

A Justice Department lawyer on Monday admitted in court that it was "fair" to say that the Trump administration's list of "likely noncitizen" voters has a high error rate.

The shocking admission from Assistant U.S. Attorney Kenneth Kao came in a hearing in a lawsuit brought last week by unions representing election workers and civil rights organizations over the Trump administration's threatened prosecution of those election workers and other related conduct.

Yep, that's the way the DOJ does business under Trump's control: You fuck up until you're forced to admit you've fucked up. Firings and forced resignations ensue. New sychophantic rubes are pressed into service. Rinse and repeat.

But what's really enjoyable about this is how fast the DOJ (and the administration overseeing it) lost at every single level of the federal court system after making this admission. Geidner has reproduced the entire timeline, which is truly a sight to behold.

At 10:10 pm (ET), the federal court handling the lawsuit denied a request for a stay of its order. This led immediately to the administration running its election interference campaign to the next two levels of the court system, hoping against hope that at least one set of judges would give the DOJ permission to fuck with midterm elections in Calfornia. No dice.

[UPDATE, 4:00 p.m. ET November 3: After a flurry of activity in the U.S. Court of Appeals for the Ninth Circuit since Judge Thompson denied DOJ's motion for a stay of her TRO, a 6-5 limited en banc panel of the appeals court issued an order Tuesday that stays Thompson's TRO pending the Justice Department's appeal of it at the Ninth Circuit.]

[UPDATE, 6:15 p.m. ET November 3: The plaintiffs are seeking full court en banc review at the Ninth Circuit, arguing, "Federalism principles of the highest order, embedded in our Constitution, require correction of this erroneous decision by the full Court."]

Meanwhile, the administration was trying to secure a parallel victory in the highest court in the land, pursuing its appeal of an earlier rejection of the DOJ's lawsuit against California — one that demanded election officials "segregate" any ballots that might be related to the DOJ's admittedly highly erroneous list of 20,000 voters.

The Supreme Court couldn't be bothered. Its rejection failed to state if it was less than impressed with the administration's lack of subtlety when it came to its own election-rigging efforts:

[Update, 6:05 p.m: The Supreme Court quickly rejected [Solicitor General John] Sauer's request on Monday evening — before any opposition was even sought or filed… There were no noted dissents.]

So, if you're following along and can read a clock, you're seeing something rather remarkable: an administration being shut down by three consecutive levels of the federal court system in less than 24 hours.

And that means California doesn't have to do any of the bullshit the DOJ demanded and then sued about. It also means the DOJ doesn't actually have any illegal election dirt on 99.9999% of California's voters and we wouldn't be shocked at all if it doesn't have any dirt on the 28 people it managed to round up during its statewide sweep.

This article is republished from The Conversation under a Creative Commons license. Read the original article.

On the evening of Oct. 11, just weeks before the 2026 midterm elections, a federal judge ordered the Trump administration to stop airing taxpayer-funded television advertisements praising President Donald Trump.

Yet two of the ads continued running past the court's deadline, including during broadcasts of "60 Minutes" and "ABC World News Tonight."

The commercials in question aired in violation of a decades-old congressional prohibition against using federal money for unauthorized publicity and propaganda. Yet Republican congressional leaders did little to stop the president's ad spending. That left outside organizations and a political candidate to challenge them in federal court.

The episode illustrates a larger development in Trump's second presidency. Across disputes involving federal spending, government agencies and military operations, the administration has asserted authority that is either restricted by law or that the legislative branch has reserved for itself.

But responsibility for enforcing those limits is increasingly falling to organizations, institutions and individuals outside of the legislative branch.

As a scholar of the American presidency and deliberative democracy, I pay close attention to how the nation's chief executive exercises power and communicates with the public.

When presidents are allowed to break rules and disregard democratic norms without consequence, I worry that citizens will grow cynical and lose faith in the idea of self-government.

As this happens, citizens may come to see the law as something that applies selectively, depending on who holds power, and lose confidence in the institutions responsible for holding their leaders accountable.

They may also begin to accept the absence of accountability as a normal feature of political life, weakening the public expectations on which democratic government depends.

Practical immunity

Congress is not the only branch of government demurring from holding the president accountable.

In July 2024, the Supreme Court provided presidents sweeping protection from criminal prosecution for actions taken in office under the "core constitutional powers" of the position.

In "Trump v. United States," the court held that former presidents cannot be criminally prosecuted for actions taken under powers the Constitution gives exclusively to the president. Along with this "absolute immunity," the justices also granted "at least presumptive immunity" for other official acts.

In their decision, the court's conservative majority extended the meaning of official conduct to the "outer perimeter" of presidential responsibilities and prohibited courts from examining a president's motives when determining whether an act was official.

The ruling only addressed criminal liability and did not curtail Congress' constitutional authority to investigate presidential conduct, control federal spending, enact statutory restrictions and, in serious cases, pursue impeachment.

Yet the Republican majority in Congress has largely declined to exercise these powers against Trump, in areas ranging from declaring war to spending money.

Their acquiescence is helping create a new form of official protection, which I call "practical immunity." In using this term, I am borrowing from the work of scholars who study abuse in the American penal system.

Those thinkers argue that legal obstacles, institutional hesitation and political incentives sometimes make it difficult to hold public officials accountable, even when the law suggests there has been illegal conduct.

That is what appears to be increasingly happening now with Congress, the courts and the president.

When Congress stops defending its laws

The television ads case illustrates the conundrum of practical immunity.

In opposing the temporary restraining order, the Trump administration's attorneys did not contest the argument asserted by the plaintiffs - Common Cause and two unions - that the expenditures violated federal appropriations law. Instead, they said the plaintiffs lacked standing to sue. They also argued that the case was moot because the advertising campaign was ending.

Those arguments failed in court. Yet the administration's defense demonstrated how elusive enforcing a law can be when Congress fails to assert its constitutionally granted role, even when the underlying legal restriction is clear. To win, a plaintiff must establish a concrete injury and persuade a judge to intervene. While this happens, the disputed conduct often continues.

In this case, the president went on spending public money in violation of the law. The GOP-led Congress, meanwhile, remained silent, even though its own role was being usurped.

Another boundary that Trump has repeatedly tested is the so-called "power of the purse", which is among the legislative branch's most important constitutional responsibilities. Presidents may spend federal money only for purposes Congress has authorized. The Impoundment Control Act of 1974 protects lawmakers' decisions against presidential efforts to ignore congressional appropriations.

In September 2026, Trump's administration canceled more than US$800 million in congressionally approved spending for immigrant services, climate change and civil rights for minority groups. The cancellation was done through a maneuver known as a pocket rescission. The administration timed its action to make it difficult for Congress to consider or stop the rescission. The Government Accountability Office - an independent agency that examines how taxpayer dollars are spent - said tactics like these violate federal law.

When this happened, a few members of Congress tried to intervene and stop the rescission, but the Republican majority blocked their efforts.

The collapse of institutional rivalry

The separation of powers divides authority among the legislative, executive and judicial branches of government to prevent too much power from becoming concentrated in any one place.

The architects of the Constitution anticipated that officials in different branches would compete to preserve their authority. In Federalist No. 51, James Madison argued that maintaining the balance established by the separation of powers depended on the ambitions of public officials coming into conflict. Members of Congress could be expected to resist presidential encroachments because those encroachments diminished their own power.

Fierce partisanship has complicated this arrangement.

Legal scholars Daryl Levinson and Richard Pildes have shown how political competition mostly now happens inside and between parties, rather than among branches of government. When one party controls Congress and the presidency, legislators may become more invested in advancing their party's agenda than they are in defending their institution's constitutional authority.

Trump's influence over Republican voters intensifies these incentives. Lawmakers who oppose him risk public ridicule, primary challenges and diminished political support. Defending congressional authority may therefore carry greater immediate - and personal - political costs than allowing presidential power to expand.

The result is a political system in which a president can act in ways that may violate federal law, undermine democratic norms and encroach on Congress' constitutional authority.

Madison's constitutional design assumed that public officials would defend the powers entrusted to them. The danger to the American constitutional system emerging in Trump's second term is that congressional leaders have begun to regard those powers as expendable, allowing the president to exercise authority under conditions of practical immunity. He faces little fear of meaningful institutional consequences.

Stephanie A. (Sam) Martin is the Frank and Bethine Church Endowed Chair of Public Affairs at Boise State University

US federal court have overwhelmingly ruled (10,000 orders and counting) that migrants are — just like natural-born citizens — entitled to constitutional rights. Among them is the right to due process. This means that detained migrants have a right to a hearing to see whether they can be held in custody until a hearing on the issue. And that issue is whether or not they can be held indefinitely.

The courts have said migrants have rights, far more often than not. The Trump administration has sought to counter this with an extremely novel interpretation of immigration laws that have been on the books for decades, claiming there's ultimately no difference between someone caught crossing the border illegally (who can be detained and deported without worrying about the Constitution) and someone who has been in this country for months or years and poses no actual flight risk, much less a threat to public safety.

Since it's not on the right side of the law, the party of "law and order" has encouraged ICE to thwart the law and/or flat-out ignore court orders. ICE has made it a habit to immediately move detainees from detention center to detention center in hopes of preventing their cases from being heard. Not only does it distance migrants from their legal representation, it keeps the venue in permanent flux, which means habeas petitions are doomed to fail because they were filed in place other than where the rights violation took place (the location of the arrest).

ICE also tends to route most detainees to large concentration camps in Texas and Louisiana because they're ultimately overseen by the Fifth Circuit Appeals Court, which has (now) repeatedly decided migrants don't have rights.

Apparently, the usual ICE bullshit isn't working well enough or often enough. Now, the administration is straight up mocking the 10,000+ decisions in favor of due process rights by deliberately targeting migrants who have been freed from detention by federal court orders.

Immigrants who have been granted release from detention after winning habeas corpus petitions in court are being re-arrested by US Immigration and Customs Enforcement (ICE).

The Guardian has identified at least 20 cases nationwide relating to people who were detained after being released on habeas corpus petition in the last year, according to court records, attorneys and media reports.

Twenty cases may seem like an immeasurable drop in the bucket compared to the nearly 100,000 habeas petitions that have been filed since Trump's assault on migrants began early last year. But it's also an obvious undercount. This determination relies on what can be gleaned from court records, which means there are likely hundreds of "re-detainees" who haven't been able to file a motion due to ICE's deliberate attempts to separate migrants from their rights and representation.

And this effort is amazingly brutal. No normal human being would consider this to be acceptable US government behavior. But this is something the psychopaths and sociopaths running the GOP are willing to do to indulge the whims of their preferred demagogue.

And in Minnesota, seven immigrants were re-detained before even exiting the Whipple federal building detention facility on the outskirts of Minneapolis, as first reported by MPR.

"I don't think being taken out of a building and directly into the arms of four or more officers complies with the court's order," Daniel Suitor, attorney to two of the men, said in a phone interview .

In Louisiana (Fifth Circuit, yo!), immigration officials were held in contempt after re-arresting a single migrant three times after he'd been granted release by federal judges. More of the same is going on in Minnesota (see above) and Pennsylvania.

Not only are migrants being re-arrested immediately after securing their release via court orders, but they are increasingly being deported to countries with which they have no ties. Cruelty on top of cruelty on top of hideous displays of inhumanity. That's just the day-to-day operations of our current administration, one completely consumed by hatred and capable of little more than inflicting whatever pain it can until it runs out of time… or victims.

We have all wanted to learn a language at some point but it's hard to get started. Some language learning tools can be complicated and very time-consuming. But with uTalk, you'll be speaking keywords and phrases in no time, and will start to see the results straight away. It helps you overcome the language barrier challenge by helping you learn real, practical vocabulary in a wide variety of languages from any device that you choose. uTalk's language programs let you listen to native speakers who are recorded in uTalk's recording studio and feature independently verified translations so you'll be able to navigate through your next vacation like a real local. Get 2 languages for $30, 6 languages for $50, or all 100+ languages for $69.97.

Note: The Techdirt Deals Store is powered and curated by StackCommerce. A portion of all sales from Techdirt Deals helps support Techdirt. The products featured do not reflect endorsements by our editorial team.

Last week, Ofcom, the UK regulator in charge of all this stuff, released its "rapid assessment of highly effective age assurance at 16." That came right on the heels of its quietly released (only because the law required it) report showing that the Online Safety Act's duty of care wasn't actually having any noticeable impact on kids' access to "harmful content."

Even as the Online Safety Act was being rolled out after years of debate, UK politicians started whining and screaming that they had to follow other countries towards banning kids from social media entirely. Ofcom is supposed to help put together a plan for such a social media ban by the end of the year, with implementation set for next spring. As part of this, Ofcom was tasked with reviewing the state of technology that can determine whether someone is under or over 16, and in doing so basically confirmed what many activists have been pointing out all along: if you're putting an explicit age limit on social media, you are requiring kids to hand over private information and put themselves at risk.

The report includes this handy chart, which basically notes that if you're trying to determine if someone is between the age of 16 and 18, you're really talking about having them hand over private information:

That first column on the left is the telling one. That's the one that says there are only four(ish) "highly effective" means of doing age assurance… and basically all of them are actually "send private info to a company." The first two are obvious: uploading a photo ID or banking info, both of which are incredibly risky, especially since you'll be uploading them to companies which are clear targets for hackers. The third one is really the same thing, because even as "digital identity services" are coming online with the goal of doing the ID part as a third party and then issuing some sort of "old enough" / "too young" token, those services are still getting that information via the uploading of private info.

But it's the last one that's the funniest. For years, age verification providers have insisted that no one would have to upload private info, because they could estimate your age just based on scanning your face. But we've always called bullshit on that. Such scanning may be able to tell the difference between a 50-year-old and a 15-year-old but it's effectively useless for determining the difference between a 15-year-old and a 17-year-old.

And Ofcom kinda admits that. It still lists facial age estimation as "highly effective," but only with that parenthetical caveat. And the full paper concedes that the technology falls apart for exactly the kids closest to the age line:

The margin of error for facial age estimation becomes more significant for the ages closest to the target age gate (16 in this case). For example, if the facial age estimation algorithm is accurate within two years for children aged 15, it could estimate them as aged 17. This could result in significant numbers of children under 16 gaining access to services or functionalities that are restricted. It could also result in a 17-year-old being denied access if they are estimated to be 15, unless they use another method. However, this is only likely to affect users who are close to the target age, as most adults offered this method would be estimated to be clearly above 16.

So, yes, if you're much older, perhaps a facial scan gets you onto a website without uploading your ID. But if you're anywhere near 16, you're basically out of luck, because Ofcom's fix for the margin of error is to kick anyone who looks young-ish over to a different method… and a different method is always going to be "upload an ID."

In order to account for the margin of error in a facial age estimation algorithm, services should implement a 'challenge age' (sometimes known as a 'buffer age'). A challenge age approach is widely used in an offline context when selling age-restricted products like alcohol in retail environments, for instance, through the retailing strategy 'Challenge 25.' In this approach, anyone who appears to the provider of restricted products to be under the age of 25 should be challenged to provide acceptable ID proving that they are over the age of 18 if they wish to buy alcohol.

Of course, requiring kids to upload their photo IDs or banking info is going to be difficult given that many kids… don't actually have photo IDs or banking info at age 16. Ofcom argues that 89% of 16-year-olds have a passport while 19% have a provisional driving license. But passports also aren't the kind of ID that kids regularly have at the ready any time they're reaching an internet service. Perhaps I'm out of touch with the youth of the UK, but I'd be surprised to find many of them regularly have passports at arms length most of the time.

The likely result is forcing more kids to get IDs at younger ages than they would otherwise need them, again creating privacy questions for those kids. Either way, teenagers are in for a lot of "your papers please!" and Ofcom says so pretty directly:

Our preliminary assessment suggests that 16- and 17-year-olds are more likely than users of other ages to need access to a photo ID document. Of the methods listed as capable of being highly effective at 16, a photo ID will likely be needed at some point in the process. This is either because the method itself requires a photo ID, because ownership of a photo ID is part of the underlying process for accessing another method (such as opening a bank account or using a digital identity service), or because a 16- or 17-year-old will most likely fall within the challenge or buffer age for facial estimation methods and be asked to prove their age with a secondary method.

In other words, what UK politicians (and plenty of "child safety activists") are demanding here is that a bunch of kids have to give up their privacy and hand over documents that can put them at risk. For their safety. At a time when there remains little to no evidence that barring kids from internet services will actually improve their safety.

One cornerstone of Elon Musk's fraud-packed SpaceX IPO prospectus includes the claim that they'll extremely quickly (as in a year or two) boost Starlink subscriber totals from 13 million to something closer to 300 million. Given the network's congestion problems and physics constraints, that's simply not realistic. Also: more users means even greater capacity problems and weird network restrictions.

To try and goose growth, Starlink has recently been making it clear they'd like to try and get into the cellular and smartphone business. But the actual plan is, as with all things Musk, a misleading, hot mess.

A key part of Starlink's plan includes plunking down thousands of meshed femtocells installed on customer rooftops alongside existing LEO satellite dishes, something one telecom analyst recently stated was "one of the top three dumbest ideas in my four decades of being in this industry."

In part because the femtocell installations are just logistically annoying and complicated, and will further will eat up capacity on the increasingly congested network (recall that Starlink customers in some areas already face "congestion fees" as high as $1500).

To goose growth Starlink and SpaceX have been gobbling up valuable wireless spectrum (with the help of Trump FCC boss Brendan Carr). They are also apparently in conversations to buy Echostar's Boost Mobile prepaid brand:

"SpaceX has held talks with telecom firm EchoStar in recent months about launching its own signature wireless service through a deal with the company, the report said, citing ​people familiar with the negotiations. The move would support the Elon Musk-led ​company's plans to build a hybrid satellite-terrestrial mobile network through Starlink ⁠Mobile, combining its satellite constellation with ground-based infrastructure to challenge US wireless ​giants AT&T, Verizon and T-Mobile."

One problem? Boost mobile is just a mobile virtual network operator (MVNO). It's primarily just customer-facing retail artifice that runs on the back of existing AT&T and T-Mobile networks; neither of which is keen to have their cozy regional telecom dominance disrupted. To actually compete, Musk and Starlink would still need to invest hundreds of billions of additional dollars in a high-powered 5G network over the next decade. And even then there's no guarantee of success.

You might recall that Boost Mobile was offloaded to Echostar as part of that doomed first-term Trump administration effort to approve the Sprint-T-Mobile merger, then hide its labor and competition impact by pretending that Dish Network would build a successful replacement 5G network. That of course never happened; the real impact was mass layoffs, higher prices and worse service for everybody.

I should note that the Trump FCC under Brendan Carr not only launched a fake "investigation" into Dish to try and force it to offload much of its spectrum to AT&T and Musk, but recent reporting suggests the Trump administration threatened Charlie Ergen with additional scrutiny if he didn't ramp up his donations to Trump's ballroom. So there's entire layers of grift going on here.

Regardless: U.S. telecom is just generally a shitty market to try and disrupt. It's extremely expensive and time consuming, and at every turn you're likely to get creatively ratfucked by giants like AT&T and Verizon, whose political influence over Republicans historically rivals Musk's. Again, even traditionally staid and bullish telecom trade mags have expressed concerns that Starlink's smartphone push is doomed.

I think this mostly all exists as a way for Starlink to hoard valuable spectrum and goose stock valuations. There may be some people at Starlink who have convinced themselves this will actually work (the fact SpaceX has been lobbying for phone unlocking rules suggests some potential earnestness), but U.S. telecom history is littered with these sorts of half-assed disruption efforts that routinely go nowhere.

I'm not prone to believe things that come out of the mouths of Verizon executives, but in this case I think Verizon CEO Dan Schulman is correct when he states it would cost SpaceX something akin to $300-$500 billion and five years build a network capable of really competing with cellular giants.

And that's just network infrastructure; you still need to convince users to sign up, provide a quality experience, and provide useful customer support (something Musk companies historically aren't interested in). While fending off the political machinations of AT&T, Verizon, and T-Mobile, who all but own the legislature in most states after a generation of lobbying and campaign contribution.

The Musk cult and our shitty corporate press will spend the next few years enamored with the fiction that "Musk is disrupting smartphones," but this mostly just reads as further stock-fluffing performance by a man with more money than sense.

It's very gross for this to be any sort of "I told you so" opening to this post, but the measles outbreak in America continues to get worse, as predicted. While last years case counts and deaths from the disease, which had been vaccinated to the point of eradication back in 2000, were a record-breaking shitshow, 2026's case and death counts have already outpaced it significantly. There are more than double the total cases of measles already compared with last year, and triple as many deaths. Through it all, RFK Jr. and his HHS agency have battled the outbreak through inaction and obfuscation of data.

The worst of the outbreaks has been in Pennsylvania this year. Unfortunately, highly infectious diseases don't respect state borders. Measles is already spilling out into neighboring states. Weeks ago, New York State declared a state of emergency due to measles infections. And, now, the state has sadly reported its first fatality from measles.

It's the first known measles-associated death in the state in over 30 years, health officials said.

"We are deeply saddened by the death of a child in Lewis County associated with the ongoing measles outbreak," the New York State Department of Health and Lewis County Public Health said in a joint statement on Monday. "We extend our heartfelt condolences to the child's family and loved ones and ask that their privacy be respected during this difficult time."

No additional information on the case was released.

We don't have any more information on the specific case or victim beyond that. But we do know that it was another child that died. Children have represented the overwhelming percentage of case counts and fatalities thus far. We also know that this death needn't have happened. Until RFK Jr. got into government, America had not had a measles fatality recorded in decades. Through vaccination programs and the resulting herd immunity, even those medically unable to get vaccinated against measles were protected. The anti-vaxxer movement undid all of that, of course, and the avatar for that movement is Kennedy.

A major concern among health professionals right now is measles cases occurring during pregnancy, given how dangerous it can be for both the person pregnant and the fetus upon birth. The CDC reportedly wanted to issue an informational alert to health professionals about the dangers of such cases and how to handle them. Unfortunately, it seems the powers running those government health agencies would prefer to hide that information from the public.

Days before the seventh preventable measles death, The Washington Post reported on Kennedy's latest push to thwart public health efforts: Career scientists at the Centers for Disease Control and Prevention—which is under Kennedy's control—were blocked from publishing an urgent health alert to clinicians regarding the risk and management of measles cases during pregnancy.

Given the previous success of vaccines and vaccination campaigns, many clinicians have never seen such cases in their careers. But measles infections during pregnancy pose severe, life-threatening risks to both pregnant people and fetuses. As such, CDC scientists wanted to issue an informational advisory through the CDC's Health Alert Network.

But the CDC Director's Office declined to provide the required approval for the alert, sources told the Post. The scientists have since moved to trying to publish the information in the CDC's publication, The Morbidity and Mortality Weekly Report (MMWR). But the peer-review process for the publication is longer than that of an advisory. Further, vaccine-related studies already approved for publication in MMWR have been suspiciously delayed, the Post reported.

Here again we see sociopathic behavior from Kennedy and its detrimental impact on the health of Americans. Kennedy doesn't want government studies on vaccine related information because, obviously, those studies don't align with Kennedy's preferred conspiracy theories about vaccines. Likewise, he doesn't want alerts going out to medical professionals about measles during pregnancy because either they put the lie to Kennedy's claim that he has this whole measles thing under control, because the advice given in those alerts would be to encourage more vaccination, or both.

There is no explanation for why informational alerts about an ongoing health crisis that were otherwise approved for publication should be blocked by someone with no medical degrees or background. This is hiding information from the public, and from doctors. It's evil.

People are getting sick and dying, most of them children. That Kennedy is prioritizing his own ego and pet projects over those lives isn't surprising, but it is horrifying. And it must be stopped.

In courts across the country, publishers and creators are suing large AI companies for training generative AI models on copyrighted works. It's easy to see these cases as a David-and-Goliath showdown between authors and publishers fighting for a cut of Big Tech company profits. But this fight is actually much larger. It affects billions of people who aren't represented in the case at all, including researchers, artists, activists, journalists, people with disabilities, and anyone who accesses and shares content online.   

That's why EFF is stepping in to defend the public's rights in these cases. Last week, EFF filed an amicus brief in In re OpenAI, explaining how the court's ruling on fair use could impact people worldwide. If fair use gets narrowed, it will impede work that benefits the public, like medical research and investigative journalism, and even limit access to privacy protection tools.

Let's be clear: creators and workers deserve support as AI changes their fields, and Big Tech companies must be held accountable for the harms they cause. But giving copyright holders more control over the web won't accomplish either of these goals. If the plaintiffs prevail in these cases, courts would have to rewrite copyright law in ways that hurt everyone except powerful media and tech companies.

Rewriting Fair Use Makes Everyone Worse Off

Copyright exists for one reason: to stimulate creativity and innovation in service of the greater public good. Unfortunately, the plaintiffs in this lawsuit are arguing to limit everyone's fair use rights, in ways that would thwart the basic purpose of copyright. The harms of shrinking fair use will go far beyond generative AI, impacting everyone.

Initial speculation that AI models would be infringement machines, churning out works highly similar to those on which they were trained, has proven unfounded. And courts have already found fair use in cases where copyrighted works were used to train generative AI models, recognizing that training uses those works for a fundamentally different purpose.

Plaintiffs in these cases argue that, even if the use is transformative, using their works to train AI is still unfair, because the AI models could generate new, non-infringing works that might compete with plaintiffs' own works. They also argue that scraping or downloading the data without permission—or in some cases, from websites containing infringing copies—was independently infringing and/or rendered any subsequent uses of that data unfair.

These arguments would put beneficial research and reporting at risk. First, if copying without permission rules out fair use, the law will never protect transformative uses that benefit us all. Researchers, journalists, and other watchdog groups gather information the same way AI companies do—but they do it to serve the public interest, and often to hold Big Tech itself accountable. For example, journalists and nonprofit researchers have used "crawling" or "scraping" to investigate potentially anticompetitive practices by tech companies, and research misinformation and hateful content posted on X. Their data acquisition techniques are virtually identical to those used by AI giants—but they lack those companies' deep pockets.

Requiring users to show that copyrighted material was lawfully obtained would also threaten everyday internet use. As EFF's brief explained, that requirement would "cast legal doubt on the billions of copies we collectively make every day through routine online activities like opening emails, clicking on an embedded link, visiting websites, sharing pictures, streaming music, and watching videos. All of this activity requires intermediate copying."

Fair use is what lets us all use the internet without being compelled to track the source of every image, video, and article that we click on, and make sure the rights are all cleared.

Additionally, while rightsholders are understandably concerned about competition from AI generated works, copyright law exists to promote new creation, not protect authors who published first. Under the plaintiffs' theory, no one could stand upon the shoulders of giants. Copyright could prevent people from writing new books inspired by themes in a favorite novel or genre, or prevent academic researchers from publishing literature reviews that distill insights from prior research—simply because readers might prefer those new publications over those that came first.

Finally, general-purpose AI models are neither designed nor usually used to create works targeted at any particular market. These models have been used to generate new insights and tools for medicine, science, journalism, art, and accessibility. For instance, an AI-powered tool called Be My AI uses Open AI's GPT models to provide visual descriptions for people who are blind or have low vision. Generative tools have also accelerated scientific research, and researchers have used these tools to develop new medical treatments and medications for conditions previously thought to be incurable.

Mandatory Licensing Would Entrench Big Tech

Some creators and publishers argue that AI developers should just get a license. But requiring permission for every work used would give the largest companies massive advantages over small competitors and impede academic research. As Professor Lawrence Lessig recently noted:

If all training requires a license, then none but a handful of tech giants will be able to afford the costs of training. That's not because authors are greedy but because the copyright system is the least efficient property system known to man.

Moreover, creators are unlikely to benefit from these licensing arrangements in many cases. Media and entertainment giants have exploited the imbalance between themselves and underpaid working artists for decades. Trying to help creators by giving them more rights under copyright is, as Cory Doctorow has written, like trying to help a bullied kid by giving them more lunch money for the bully to take. Artists aren't getting good deals from record labels or publishers now; there's no reason to think these companies, which already own the copyrights, will treat artists more fairly once they control the data that powers AI.

The Future of AI Is Too Important to Leave to Big Content or Big Tech

AI technologies will affect all of us, and copyright law is ill-equipped to address many of these consequences. The rules governing these technologies should be rules that serve the public—not just the small number of companies that can afford to litigate copyright battles for decades, or settle for billions.

Republished from the EFF's Deeplinks blog.

Support us on Patreon »

For the past few weeks on Techdirt, we've been publishing The Metric Is Not The Mission, a ten-part series by internet governance and digital policy expert Konstantinos Komaitis which is now available as a single, complete paper. This week, Konstantinos joins the podcast to talk all about his argument in the paper and why it's a necessary part of understanding big tech.

You can also download this episode directly in MP3 format.

Follow the Techdirt Podcast on Soundcloud, subscribe via Apple Podcasts or Spotify, or grab the RSS feed. You can also keep up with all the latest episodes right here on Techdirt.

So, it's come to this. The good ship USS Revenge Prosecution is sinking. And the deserting rats are deciding which rats should desert first, apparently based on nothing more than the current state of the organizational chart.

Two people figured heavily into the disastrous attempt to prosecute a handful of Chicago residents for engaging in protected First Amendment activity. The DOJ, urged on by the DHS and Trump himself, wanted to ring up hundreds of people on federal charges for engaging in anti-ICE protests. By the time the DOJ got things in front of a grand jury, the number of defendants had dwindled to… six.

They were the so-called "Broadview Six." And while the DOJ — represented locally by US Attorney Andrew Boutros — thought it had secured a handful of wins via the grand jury sign-off on the charges, it only took an initial visit to federal court to find out how wrong it was.

Here's what Boutros did to salvage what was left of his reputation when everything suddenly went disastrously sideways in from of federal judge April Perry:

Boutros announced the decision to dismiss the remaining charges in court following a closed-door meeting over redacted grand jury transcripts. He told U.S. District Judge April Perry he was unaware until recently of the alleged misconduct, including a prosecutor meeting with a grand juror outside proceedings and other jurors who disagreed with the case being dismissed and prevented from participating. Boutros did not dispute the allegations, saying the conduct was upsetting and the reason the case was being dismissed.

"Not good enough!" said the righteously outraged court, which demanded more answers from the DOJ, as well as raising the specter of sanctions if said answers weren't acceptable:

Although I am not going to prejudge the issue without a hearing, I will say that I was incredibly shocked by the redactions that were made. I have read hundreds, if not thousands, of grand jury transcripts involving prosecutors who are the most junior of prosecutors to several U.S. Attorneys who appeared before the grand jury. I have never seen the types of prosecutorial behavior before a grand jury that I saw in those transcripts.

And this was all over a handful of federal misdemeanor charges, after Boutros and his subordinates (1) did shady shit with a grand jury, and (2) couldn't actually get anyone to bite on felony charges against the "Broadview Six."

US Attorney Boutros still has his job, despite being this debacle's ringleader. The person who has been permanently turfed — former special agent in charge of the Chicago field office, Douglas DePodesta — wasn't actually involved in the failed prosecution. But he was the one who paid the price for this DOJ failure, as WBEZ Chicago reports:

When charges were dropped against the "Broadview Six" amid revelations of apparent prosecutorial misconduct, "Boutros began a campaign within the Department of Justice to blame DePodesta for the unraveling of the case." Boutros' complaints to superiors in Washington ultimately led to DePodesta's ouster, the lawsuit says.

More succinctly put:

"I was forced to leave my job," DePodesta said. "I was ordered to leave my job."

DePodesta's ouster is likely the result of a couple of things. First, Boutros desperately needed a patsy to take the hit on his behalf so he could continue his disservice to his office and his country. Second, DePodesta was a holdover from Christopher Wray's stint as Trump's FBI Director — an FBI director Trump only likes slightly more than James Comey.

The lawsuit [PDF] alleges that DePodesta's rights (First, Fifth, and others) were violated by this forced resignation. It hopes to secure not only back pay and other damages, but also a formal statement from the Trump administration admitting to violating the former FBI SAIC's rights.

Most likely, none of that will happen. But while it works its way to not happening, we can hope that it will result in a bunch of discovery that exposes more of the inner workings of Trump's DOJ-enabled revenge operations.

But before we give DePodesta the ol' "Comey is my homey" treatment, let's not forget that he wasn't exactly opposed to Trump's anti-migrant/anti-civil rights overtures. In an interview with the Chicago Tribune shortly after his elevation to the top spot in the FBI's Chicago office, DePodesta not only echoed a bunch of dumb fentanyl-related talking points, but also affirmed his loyalty to the Bigoted New World Order Trump sought to create by turning every federal law enforcement agency into little more than DHS deputies:

DePodesta said the Chicago FBI, which has more than 1,000 total employees, has partnered locally with the Department of Homeland Security and Immigration and Customs Enforcement, or ICE, embedding some of the FBI's 450 agents on the street "helping them effect arrests."

The targets, DePodesta said, "are people that have removal orders that have a criminal history — murderers, rapists," adding that the effort was "ongoing" and had no timetable for a conclusion.

"It is not a surge; it is a sustained effort," DePodesta said. He said no new agents have been added to fulfill the FBI's role in immigration enforcement — instead, agents from other squads have rotated and worked overtime.

DePodesta was onboard with rerouting FBI resources to help with mass deportation efforts. His contribution here contributed to the events that led to the bullshit charges against anti-ICE protesters. And, until we see something in forthcoming documents that states otherwise, there's little reason to believe DePodesta wasn't, at best, just saying nothing as the DHS and ICE utilized shared resources to surveil protesters and use whatever they could find to generate criminal charges.

So, while DePodesta wasn't involved at all with the extremely questionable prosecutions the DOJ engaged in here, he also doesn't seem to have done much to keep himself or his direct employees from becoming the very tools of vengeful oppression he's suing about now that he's out of a job.

Again, some judgment can be withheld here because all the facts aren't available. But I'm willing to bet that once they are made public (assuming the administration doesn't just write a check to DePodesta and tell him to walk away from this), it will not only expose the DOJ's complete subjugation, but will likely show DePodesta's agreement to contribute to the mass deportation effort won't see him emerge from this lawsuit as an anti-MAGA hero.

The Luminar Web and Mobile Bundle provides new users with full browser-based access to Skylum's complete editing toolkit, along with full access mobile app access for streamlined cross-application editing. The bundle also includes an instructional video teaching you a complete, practical photo editing workflow in Luminar for Mobile from photographer and filmmaker Nikolaj Green, in six video lessons that run about 50 minutes in total. There's also a macro-photography eBook. The bundle is on sale for $59.

Note: The Techdirt Deals Store is powered and curated by StackSocial. A portion of all sales from Techdirt Deals helps support Techdirt. The products featured do not reflect endorsements by our editorial team.

While Steve Huffman and Alexis Ohanian have tried to downplay Aaron Swartz's role in "founding" Reddit, he was still an instrumental part of the company's early journey. His "infogami" project was rolled into Reddit, and the founder credit was a part of the deal. Part of the point of infogami was to support open access to information — which was Swartz's lifelong mission. The code behind infogami is considered instrumental in the launch of the Internet Archive's Open Library.

So it's a real shame that Reddit has been doing its best to hammer the final nails into the coffin of the open internet. The latest, of course, is Huffman announcing that Reddit is moving to a "logged-in-only" experience.

Reddit will start testing the login requirement in "select," non-US markets in Q4. The company will use those tests to decide what a logged-in-only Reddit will look like "globally," per Huffman.

"Pretty much all of the major ad-based platforms have gone through this transition. We are excited and moving quickly," the CEO added, noting Reddit's intent to become a "direct daily habit for every Internet user."

Huffman admits that he's doing this in part because traffic from Google had declined massively, which can be (understandably) blamed on Google's AI summaries on search results.

"During October, we have seen that search traffic has dropped by more than 1 million referrals on average per week. If this rate of decline continues, current trends would suggest that total DAUqs [or daily active unique users] could fall in Q4, by a low-double-digit percentage from our current base of 130 million DAUqs," Huffman wrote.

People have long known that for certain searches on Google you got better results by adding "reddit" to the keywords you searched for, so that Reddit results got prioritized. At the time, people argued that this showed Reddit was becoming a "better search engine than Google," and that the company had room to build on that and maybe even compete with Google.

It hasn't really turned out that way. Over the last few years, Reddit and Huffman seem to be seeing how much of the open web they can cut off in pursuit of locking things into silos they can charge for. You may recall three years ago when Reddit cut off a bunch of third party developers (some of whom made it possible for subreddit mods to do their job) all in pursuit of a paid API with the belief that by locking down its tools, the company could profit off the content of its users (despite Reddit not holding the copyright on their content, and not sharing any of the money it would make back with the content creators).

Reddit signed a big $60 million/year API deal with Google, and that should have been seen as the beginning of the end. Now, instead of working on making its site better for users, Reddit was deep down the enshittification curve of making it worse for users to support its business partners who paid the bills. Last year we called out the company for its horrifyingly problematic lawsuit against SerpApi, in which it claimed that because SerpApi made Google search results available to third parties as an API, it violated the DMCA 1201's anti-circumvention clause.

The whole thing made little sense, because it was difficult to see how Reddit could claim standing with a straight face. It was complaining about a service that operated on Google, not Reddit, that made it possible that others could find links to Reddit posts, with content from Reddit's users (again, not Reddit's own IP). That case is still ongoing, but it was yet another attack on the open internet by Reddit, as it seeks to wall off the openness in pursuit of a form of scarcity it can charge extra for.

And, certainly, part of this new effort is in the hopes that the more they lock up their own content, the more they may be able to cut licensing deals.

Cutting off logged-out visitors is just the latest step. In Huffman's writeup, he admits that this is an anti-open internet move, which he claims is new to him:

And I don't know about you, but as someone whose life has been defined by the open internet, it's a big change for me. But I've come to the conclusion that this is the right decision for our platform and the right decision for our business.

But, again, Huffman has been moving squarely in this direction for the past three years, so I wouldn't say it's entirely accurate to say this is a "big change" for him.

Huffman and Reddit think this will lock more users into Reddit, but I would bet that long term it will do the opposite. Just as locking down the API to cut a deal with Google didn't bring more traffic to Reddit itself, locking out logged-out readers seems likely to push people toward alternative sources of information rather than into Reddit's silo. Huffman argues that logged in users are more profitable and valuable for the company, so is betting that this strategy will trade drive-by visitors for more valuable logged-in users, but I wouldn't take that bet.

There is some extra irony here as well, given that when the Australian government began pushing its "under 16 social media ban" and included Reddit in the list of social media, we were told it wasn't really a ban, because you could still enjoy Reddit without an account. You just couldn't have an account if you were under 16. But now, if you don't have an account, you'll be much more limited in what you can access. There apparently will still be a front page that you can read without logging in, and they want to build some sort of "incognito" mode, but the site is clearly moving to a siloed logged-in experience.

It's also why I'm concerned about efforts to require a license for all scraping, which is exactly the logic behind Reddit's lawsuit. That would also be the death of the open internet. The more we wall off the internet to fight AI, the more we lose the wonders and value of an open internet. If people want to support a true open internet, where anyone can build, we're going to have to figure out a better overall system than just putting up walls and saying everything must be licensed.

Meanwhile, I'm extremely biased on this, but I will note that ATproto Spaces has been in alpha for a couple months now and should be more widely available soon. And one thing it will enable developers to do is to create Reddit-like spaces on the open internet that no central entity can control. As the old web starts closing down, perhaps we can see the rise of a new open web, where the users are in control, rather than giant companies on the enshittification curve.

For more than a generation, U.S. media academics have warned anybody who'd listen that the death of your local newspaper and the steady consolidation of local TV broadcasters was creating either "news deserts," or local news that's mostly just low-quality infotainment and agitprop.

Despite claims that the "internet would fix this," doing quality local journalism just wasn't profitable enough for the dipsy brunchlords that fail upward into positions of prominence at most media companies, so the internet… didn't. As a result, all sorts of terrible stuff has rushed to fill the vacuum where quality local journalism once existed.

That includes not just your usual collection of propaganda like Fox News, Tucker Carlson, Sinclair Broadcasting, etc., but also so-called "pink slime" fake local newspapers that try to look like a real local paper, but are basically just partisan agitprop.

Tow Center for Digital Journalism at Columbia University did the definitive study on the subject in 2022, and found that there were 1,200 bogus local news outlets around the country, all feeding gullible readers a steady diet of misleading bullshit. In the years since, such fake newspapers have outnumbered real ones. They're often even better funded.

This partisan ratfuckery historically was heavily dominated by Republicans. But in recent years Democrats have gotten in on the action. And increasingly as voters turn to AI to inform them about upcoming elections, AI chatbots are leaning on these fake pink slime newspapers as reliable sources of local information:

"As a citation, ChatGPT linked to a website called the Kalamazoo Beacon. The Beacon labels its articles as "news articles" in the website's source code. The outlet also describes itself in the site's source code as "Reporting on Kalamazoo, southwest Michigan, and the issues shaping lakeshore communities."

But the Kalamazoo Beacon has only published 16 stories, all about [Michigan Democrat Democrat Sean McCann], his policy positions and his record as a state legislator, NPR's review shows."

When the "both sides" press covers this pink slime phenomenon, they sometimes try to imply that this kind of stuff is happening perfectly symmetrically among both major parties (as this Financial Times story does). But NPR's own previous reporting (which this newer piece curiously fails to mention) indicates that Republican pink slime efforts outnumber Democrats by around ten to one.

In this newer case, nobody can actually seem to track this new smattering of pink slime newspapers to an actual source:

"It's unclear who operates the websites. NPR was unable to find or reach the site operators. One of the Democrats promoted by the network, JoAnna Mendoza of Arizona, said she wasn't behind the websites and doesn't know who is. The Democratic Congressional Campaign Committee, the campaign arm of the House Democrats, denied involvement in a phone call with NPR. Other candidates the websites promoted, including McCann, as well as other Democratic super PACs that support the candidates, including American Bridge 21st Century, Tomorrow Together and House Majority PAC, did not respond to NPR's requests for comment."

Regardless of the party of origin, the tactic represents a very ugly problem in a country increasingly being devoured by artifice and disinformation. There's an active extraction class effort to not only undermine sources of expertise (journalism and academia), but to "flood the zone with shit" to disorient Americans (or their AI chatbots) that go hunting for reliable information.

The death of local news divides the public, can actively shift electoral outcomes, reduces electoral participation, makes a populace more susceptible to propaganda, and can even increase municipal borrowing costs (less local truth-to-power journalism means less oversight of waste, which results in more waste, which results in higher interest rates for municipal bond lenders).

Journalism ideally is the glue that contributes to a healthy civic nervous system. We've lost that tether, and therefore (quite intentionally) our sense of a collective, shared reality. You may have noticed a few… repercussions here and there.

There are ample paths out from inside the problem (better media literacy training, enforcing antitrust law, funding public media, restoring media consolidation and media diversity ownership requirements, creating better funding models for real journalism). And while there are organizations spending a lot of time working on the problem (see: the creation of state "civic information consortiums") they're consistently out-financed and in desperate need of additional support.

As I continue to say, you may be done with the measles, but the measles is not done with you, nor the entire country. Coinciding with the profane installment of RFK Jr. as Secretary of Health and Human Services, 2025 saw the largest number of cases of measles in America in 3 decades. After decades of preaching that established vaccines such as the MMR vaccine were dangerous, caused autism, or were simply not tested enough for safety to be in use, Kennedy oversaw the fruits of his labor as Americans, particularly children, got sicker. Through it all, Kennedy confronted this reality with conspiracy theories, or claiming he had this all under control, or with suggestions that real problem is dirty immigrants infecting Americans.

You will note that none of this represents government actually doing anything about the measles outbreaks that are now all over the place in America. Due to this inaction, we eclipsed 2025's case count all the way back in July. And because of the nature of highly infectious diseases, we have now reached a case count that has more than doubled that of 2025.

New data from the U.S. Centers for Disease Control and Prevention shows there have been more than 4,580 measles cases reported in 2026 across the U.S. as of Oct. 22, which is more than double the number for the whole of 2025.

Something like a quarter of the case counts have come from Pennsylvania alone. Some of the talk from the administration and others is using that fact to try to paint this as a single-state problem, but it absolutely is not. Nor is this an "Amish problem," which is a claim stemming from the fact that the highest case counts are occurring in counties with large Amish populations, that typically have lower vaccination rates.

This is important: that's just not how infectious diseases work. A problem for one county is a problem for the surrounding counties. A problem for one state is a problem for the country. The American people move and travel freely, and allowing a disease as insidious as measles to fester means it will spread.

Vaccination remains the best protection against this disease, even as some breakthrough cases continue to occur among the vaccinated. And it seems that the public is starting to wise up to the fact that Kennedy's warning against vaccines is bullshit, particularly those closest to the outbreaks.

One of the best ways to protect oneself from getting the measles is vaccination, according to health officials. Two doses of measles, mumps, and rubella vaccine known as MMR, is nearly 97% effective against measles, according to the CDC. One dose is about 93% effective.

In Pennsylvania, the health department said healthcare providers have seen more residents interested in getting vaccinated for measles. Across the state, more than 39,000 measles vaccines were administered during the first three weeks of October - higher than an average of 25,000 vaccines administered across the state in one month, the health department said.

Getting to the other side of all of this will not be done quickly. It's going to take time. Vaccination rates in America continue to fall overall, which is going to be a headwind against getting measles in America under control. Perhaps changes will be made after the midterms that will lead to actual government action on this problem. Still, it will take time.

Sadly, a good first goal here would be for there to be fewer measles cases in 2027 than what will be a record-breaking total in 2026. Given that measles was declared eliminated in 2000, a goal like that is embarrassing. But at least it would be a start.

But for that, we need federal leadership that just isn't in place at the moment.

Walled Culture has just written about the possibility that recent AI breakthroughs might accelerate a move to open access, at least in the field of mathematics. It would be particularly welcome against a backdrop of relatively slow progress in the open access world. That is not because the need for different business models that provide free access for all to published research is less, but simply because the obstacles and difficulties faced by researchers remain largely unknown to the wider public, and therefore removing them is not a priority.

To the outsider, the world of academic publishing seems to be flourishing; and it is, for publishers, but it remains a source of friction and frustration for the people actually writing the papers that are published. The wider dynamics within the world of academic publishing remain opaque, particularly concerning the costs and benefits. Two new independent studies from the University of Sydney Business School may help to change that. The first, entitled "The Cost Of Knowledge" notes a remarkable omission in the discussions about academic publishing and the move to open access:

Debates about the economics of scholarly publishing typically focus on subscription prices, article processing charges, publisher revenues, and profit margins. Much less attention is paid to the costs incurred in producing the research that makes scholarly publishing possible.

Publishers love to talk about how much it costs to produce academic journals, and why the charges for publishing have to be so high, but they gloss over the fact that publication is merely the end stage in the much longer and more complex process of conducting and reporting research. The new University of Sydney Business School research seeks to estimate how much that work before publication typically costs, in order to provide context for the publication fees. The researchers identified eight key stages of article production where costs could be estimated:

ideation and conception; grant seeking and approval, where applicable; ethics approval; performance of the study and data analysis; writing and manuscript preparation, up to first submission; peer and conferencing feedback; journal editorial work; and peer review and revision cycles. Rather than costing these stages as discrete, sequential blocks, we represent them within a single integrated estimation record that recognises the overlaps between them, the resources they share, principally researcher time, but also institutional infrastructure and administrative support, and the cumulative effect of revising as a manuscript moves iteratively between stages.

The final result of applying that methodology is striking:

Drawing on the ingredients method, this paper attempts to construct a fuller account of that process, tracing the resources and associated costs consumed from ideation through to acceptance. While our estimate of approximately US$70,000 is necessarily conservative and open to refinement, it points to a substantial investment of labour, infrastructure, governance arrangements, and public resources that are essential [for] the production of a published article long before publication itself becomes possible.

To place that figure in context:

Grossman and Brembs (2021) put the cost of [the service provide by academic publishers] at around US$400 per article. Set against our estimate of approximately US$70,000 to produce a social science article, the publishing function appears to represent less than 1% of the total resources invested in creating the research. Even using Grossman and Brembs' upper estimate of approximately US$1,000 per article, publication costs account for only around 1.4% of total production costs. This comparison highlights a central tension in contemporary scholarly communication given that after substantial public and institutional investment in producing research, access to the resulting knowledge is frequently restricted through subscription-based publishing models.

The implications of that relatively small figure are important:

Viewed in this way, the estimate draws attention to a structural asymmetry within contemporary scholarly communication where the production of knowledge depends heavily upon publicly funded institutions, publicly supported research, and academic labour, and this provides a largely unaccounted for subsidy to private sector publishers, underwriting at least some of the financial returns they enjoy

That is, with every paper that appears in their titles, academic publishers are generating significant profits not only thanks to free academic labor in the form of peer review, but also thanks to long-term and massive levels of public funding. Their ability to free ride on such invisible subsidies is never acknowledged by publishers, which like to claim all the credit for making papers available, as if they were benefactors underpinning the entire system of research. In fact, by imposing high and unjustified publication costs on academic institutions and their funders, publishers act as a serious obstacle to the dissemination of publicly-funded research, denying people their right to access the academic work they pay for.

The results of that "structural asymmetry" are explored in the second piece of research from the University of Sydney Business School, with the self-explanatory title "How Profitable are the Big Four Publishers" - the "Big Four" being RELX (Elsevier), Informa (Taylor & Francis), Springer Nature and Wiley:

The evidence shows that academic publishing has been an exceptionally resilient and profitable business. Despite periods of major economic disruption, including the global financial crisis (2008) and the Covid-19 pandemic (2020-2022), all four publishers remained profitable throughout the period examined. For the diversified firms, academic publishing consistently generated the highest profit margins within their business portfolios.

…

Taken together, the findings suggest that the Big Four are not only securing significant financial returns from scholarly communication but are also extending their influence across the wider research ecosystem. The report raises questions about the distribution of economic value, control of research infrastructure and the governance of publicly funded knowledge.

The two new reports are unlikely to change the state of open access quickly or dramatically. But their value lies in the fact that they confirm - with data - that the current academic publishing system is a terrible deal not just for researchers and institutions, but also for the governments and public which fund it.

Follow me @glynmoody on Mastodon and on Bluesky. Originally published to Walled Culture.

Last week Cathy wrote about the amicus brief we filed in the big multi-district litigation (MDL) that consolidated a bunch of different lawsuits claiming that OpenAI and Microsoft infringed the plaintiffs' copyrights by training AI models on their works. A lot of people and groups offered proposed amicus briefs. For boring procedural reasons, unlike in most cases, you couldn't just file the briefs, but rather everyone had to ask permission from the court, which meant that each attempted filing involved multiple filings, and many potential amici used different processes to try to file their briefs. Oh, and the docket is already at almost 2,500 entries. It's a mess.

So, if you want to review all of the proposed amicus briefs, I've embedded below a tool to look at all of the briefs. They are sorted by which side they support (defendants, plaintiffs, neither), and then if you open up each section you can see who the amici are, a link to their brief, and a one sentence summary of their argument. In case you are wondering, yes, this was created using Claude Code and the CourtListener MCP, asking it to go in and find the actual proposed briefs from the docket, to organize them, and to summarize them. And yes, not surprisingly, it didn't get everything right on a first pass. The version below is the third version, human reviewed, after it had trouble reading some of the files and got confused on who some of the amici were actually supporting.

Anyway, we already have a whole post on our own brief, but a few others I'll call out quickly. While it's no surprise that the usual copyright maximalists filed briefs supporting the plaintiffs, what's interesting is how many other copyright holders filed in support of the defendants like OpenAI. My favorite one is from The Authors Alliance, a group representing authors, but who recognize that copyright maximalism and hating on fair use is bad for authors as well. Their argument is to call bullshit on the whole idea of "market dilution" being a harm implicated by copyright, noting: "Market Dilution is just a fancy term for competition, and copyright law supports competition."

Other copyright holders who weighed in on behalf of innovation were independent creators Sarah Rose Siskind and Michael Wall. They argued that contrary to the claims of plaintiffs, AI is helping creators as a tool that assists them in being creative. Indeed, they highlight how it has helped them compete with big-name creators backed by major industry players. The brief from the Chamber of Progress and the Abundance Institute also included the incredible multi-Grammy winning drummer Trevor Lawrence Jr., also talking about how important AI has been in helping musicians like him create new music.

There were a number of other industry groups and civil society orgs, along with a bunch of academics. EFF details how a ruling for the plaintiffs would do real damage to research. Library groups and the American Council of the Blind teamed up with tech trade groups CTA and BSA to highlight the many important benefits of AI tools. And then there are the academics. Copyright experts Pam Samuelson and Chris Sprigman (both former Techdirt podcast guests) also call out the nonsense "market dilution theory." Brian Frye teamed up with journalist Ron Suskind to explain how training is transformative and how AI is useful for journalists. Professors Rebecca Tushnet and Edward Lee — both well known copyright experts as well — also explain transformative use and how silly the market dilution theory is.

There are some briefs that support "neither" party (technically the Frye one supports neither, but the argument seems to strongly support the defendants) including the Wikimedia Foundation's brief. It focuses on applying the (problematic) precedent from the weird Andy Warhol fair use case a few years back, arguing that under Warhol, courts need to analyze each use separately, rather than making a blanket assertion that [waves hands around vaguely] "all this is infringing." It does say that regular AI training should be fair use, but argues that RAG training (basically, AI on top of a corpus of data) probably isn't fair use.

Not surprisingly, I'm less impressed by those siding with the plaintiffs. To me they trot out the well worn and debunked tropes of copyright maximalists of past fights. Claims that "if this isn't infringing no one will ever create again" are rampant and just silly. The oddest one may be actor Ashley Judd's. It basically argues that AI will kill journalism (citation desperately needed) and that if that had happened a decade ago, there wouldn't have been journalists left to blow the whistle on Harvey Weinstein and the entire #metoo movement might not have happened. I honestly had trouble following how any of those things connect, especially since journalists use AI tools all the time, and it hasn't made them disappear.

There's a similar one from everyone's favorite documentarian, Ken Burns, who argues that AI will kill the market for historical documentaries. Again, I don't see how anyone gets there, though. First of all, the market for documentaries has grown massively over the past decade, and they're now often treated as equals of feature films, something that would have been unthinkable not long ago. And literally this amicus brief is filed the same week people are debating if the new Elizabeth Holmes documentary has a shot at winning the Oscar for Best Picture (rather than best documentary). And I know (personally!) documentary filmmakers who are saying that AI tools are often making it easier to create stronger documentary films (and no, not by generating AI slop, but to help with some of the more annoying production tasks). Then there are all the expected ones from the RIAA, Copyright Alliance, newspapers, the publishers… who say the same tired "oh no the sky is falling, creators will never create again" bullshit.

The one other surprising (and disappointing) filing on behalf of the plaintiffs was the brief from PEN America, the Protect Democracy Project, and Reporters Without Borders. I am a fan of all three organizations and find it incredibly awkward and odd that those organizations in particular would strike an anti-free speech stance by siding with copyright maximalists seeking to lock up knowledge. They give a highly misleading history lesson on the Progress Clause, suggesting that the First Amendment's Press Clause and the Progress Clause work together in a way that justifies locking up knowledge for the benefit of the press. They're right that the two clauses fit together, but they get the relationship exactly backwards: both exist to expand the spread of knowledge, not to restrict it. It's just… odd.

Anyway, you can find the links and summaries to all those and more below… This case still has a long way to go and, no matter how it's decided, will almost certainly involve many years of appeals. But the copyright battle lines are being drawn, once again.

---

Proposed amicus briefs on the RECAP docket through ECF 2462, by the side each supports. Click a section to expand it.

■ 22 for defendants  ·  ■ 26 for plaintiffs  ·  ■ 4 neither/mixed  ·  ■ 1 brief not yet located  ·  53 total

Supports defendants (22) ECF 2112-1 · Oct 16 Kevin Kelly AI speeds up the creation of new works without harming plaintiffs, and its public benefits are vast. Copyright isn't a shield against disruptive technology; training is fair use. ECF 2313-1 · Oct 29 Public Knowledge Copying for training is fair use. Fair use also allows reasonably necessary copying to reach unprotectable facts, including facts about creative works. Supports defendants "in part." ECF 2335-1 · Oct 29 Engine Advocacy et al. (startups) Startups build products on fine-tuned foundation models, which shows training is a general-purpose fair use. Licensing costs would lock startups out of AI. ECF 2369-1 · Oct 30 Floor64 / Copia Institute Copyright can't and shouldn't be used to throttle AI training. Doing so would conflict with the Progress Clause and the First Amendment, and copyright grants no right to foreclose training. ECF 2370-1 · Oct 30 A.I. Progress, Inc. (AI developer coalition) Generative AI is now everyday infrastructure. Training is "quintessentially transformative," and a ruling against fair use would hurt the broader economy. ECF 2380-1 · Oct 30 Profs. Pamela Samuelson & Christopher Jon Sprigman Reject the market dilution theory: the empirical studies behind it are flawed, and any collective harm is for Congress to address. At minimum, require proof. ECF 2381-1 · Oct 30 Prof. Joshua Gans Economic framing aimed at the news cases: lost sales aren't by themselves copyright harm, dependence on any single work in a large dataset is unproven, and later licenses don't prove a market. Asks the court to reject the News Plaintiffs' arguments. ECF 2384-1 · Oct 30 Consumer Technology Ass'n, American Council of the Blind, American Library Ass'n, Ass'n of Research Libraries, BSA Catalogs public benefits of LLMs across accessibility, agriculture, law, transit, manufacturing and independent creators. Grant summary judgment to defendants. ECF 2407-1 · Oct 30 Electronic Frontier Foundation Fair use turns on the ultimate use, not how copies were acquired. Good faith isn't required, and competition isn't market harm. Plaintiffs' theories would threaten web scraping, research datasets and search. ECF 2429-1 · Oct 30 CCIA, NetChoice, TechNet Training is non-expressive intermediate copying to build a new tool (Google Books, Oracle). Market dilution isn't cognizable harm, and owners can't create a market by demanding licenses for a transformative use. ECF 2439-1 · Oct 30 Khosla Ventures VC perspective: a ruling against fair use would entrench incumbents, chill investment and cede AI leadership abroad. Grant OpenAI's motions. ECF 2449-1 · Oct 30 Gregory Leppert Libraries rely on general-purpose models for preservation and access at scale. Anyone who wants Game of Thrones goes to the library rather than rebuilding it from AI output. ECF 2459-1 · Oct 30 Former Sen. Rick Santorum & former Rep. Rick Boucher Congress made fair use flexible. AI training continues the tradition of non-expressive technological uses, and fair use underpins US tech leadership. ECF 2461-1 · Oct 30 Midjourney, Runway AI Training is designed to enable new expression. Detailed examples of AI tools expanding creativity for amateurs and professionals make it a "quintessential transformative fair use." ECF 2462-1 · Oct 30 Trevor Lawrence Jr., Chamber of Progress, Abundance Institute Extracting non-expressive patterns is transformative, and working musicians' experience shows generative AI enables human creativity. Takes no position on how the works were acquired. ECF 2328-1 · Oct 29 Authors Alliance Market dilution, as framed in Kadrey, is just competition. Factor four covers only substitution by uses that take protected expression. Single issue; takes no position on anything else. ECF 2283 · Oct 28 Prof. Brian L. Frye Training is transformative, and requiring licenses would entrench the biggest incumbents. Styled "neither party" but argues fair use. The court dismissed his earlier filing (2282) for not requesting leave (Order, Dkt. 2412); this one may have the same defect. ECF 2308-1 · Oct 29 Prof. Matthew Sag Training is highly transformative under Bartz, Kadrey and the copy-reliant technology cases. Transformativeness is judged by realistic outputs, and questionable sourcing belongs in factor four. Says it "supports neither party" but argues fair use. ECF 2378-1 · Oct 30 Foundation for American Innovation, David Shedd, Gen. Michael Groen, George Barnes Under Oracle, the court should weigh the national-security benefits the record ties to the challenged training as part of factor four. Narrow ask that in practice helps defendants. ECF 2394-2 · Oct 30 Profs. Rebecca Tushnet & Edward Lee Scaling training data with intermediate copies is highly transformative under factor one. Plaintiffs' market dilution theory rests on non-infringing outputs that share only genre or facts, which are uncopyrightable, and the Copyright Clause forecloses it. ECF 2403-1 · Oct 30 Sarah Rose Siskind & Michael Wall (independent creators) A comedy writer and a composer who license their own work describe using AI for business operations and ideation. They argue it lets independent creators compete with better-resourced rivals and helps them make more work, not less. ECF 2447-1 · Oct 30 Computer scientists Mike Knoop, Prof. Daniel Lowd, Prof. Douglas C. Schmidt Technical brief: training fits numerical parameters and is not a system for storing and retrieving works. Memorization is limited, model-specific and reducible, and research shows OpenAI's safeguards block substantial extraction, which supports a different purpose under fair use. Styled in support of OpenAI but says it takes no position on the ultimate merits. About 130 of the 160 pages are an appendix of CVs. Supports plaintiffs (26) ECF 2304-1 · Oct 29 Association of American Publishers, Copyright Clearance Center, Independent Book Publishers Ass'n, STM Training uses expressive content for the same purpose as the originals and produces substitutes, so it isn't transformative. Unlicensed training harms book markets and the emerging collective AI-licensing market. ECF 2306-1 · Oct 29 Center for American Rights OpenAI's broad fair use theory would undercut the administration's AI/IP framework. Protecting IP is a national-security issue, and a pro-fair-use ruling would give foreign adversaries legal cover for IP theft. ECF 2325-1 · Oct 29 Digital Content Next AI tools that summarize publishers' content without sending users to the source cut off traffic, subscription, ad and licensing revenue. That is an existential threat to digital publishers. ECF 2332-1 · Oct 29 AFM, SAG-AFTRA, WGA East, WGA West A flood of AI-generated works is the "substantially adverse impact" Campbell anticipated. The unions already bargain licensing for new technologies, so the court should reject a categorical fair use exemption. ECF 2359-1 · Oct 29 Copyright Alliance After Warhol, the older transformativeness cases defendants rely on carry little weight, and Kadrey and Bartz are distinguishable. Knowingly using pirated sources, and an existing licensing market, weigh against fair use. ECF 2377-1 · Oct 30 Open Markets Institute Mass copying is unfair competition against publishers. It also undercuts AI rivals that pay for licenses and could kill the emerging licensing market. Also misfiled under ECF 2372. ECF 2373-1 · Oct 30 News/Media Alliance, National Newspaper Ass'n, National Press Photographers Ass'n, ProPublica Copying news to build products that supersede it isn't fair use, relying on the Third Circuit's Thomson Reuters v. ROSS ruling (Sept. 2026). Factor four: harm to online news and AI-licensing markets. ECF 2382-1 · Oct 30 Minderoo Centre for Technology & Democracy, University of Cambridge Opposes defendants' summary judgment motions. What the models retain from books is a disputed material fact. Training is cumulative, so the copies aren't "intermediate" under Sega and Connectix. ECF 2386-1 · Oct 30 AI Coalition for Data Integrity, WGA East & West, Nat'l Ass'n of Voice Actors, Digital Media Licensing Ass'n, Transparency Coalition AI, Ass'n of Medical Illustrators, Ass'n of Independent Music Publishers, Music Workers Alliance, Microstocksolutions Answers the DOJ's national-security argument: security agencies want documented training-data provenance, which unlicensed scraping can't provide. Foreign rulings such as GEMA v. OpenAI undercut the claimed US advantage. ECF 2388-1 · Oct 30 First Amendment scholars: RonNell Andersen Jones, Jack Balkin, Erin Carroll, Alan Chen, Heidi Kitrosser, Jane Kirtley, Burt Neuborne, Jonathan Peters, Victor Pickard, Francesca Procaccini, Sonja West The press serves constitutional and democratic functions that depend on a viable business model. AI chatbots free-ride on reporting and erode that model without doing what the press does. ECF 2393-1 · Oct 30 RIAA, NMPA, A2IM, ASCAP, BMI, Recording Academy, Songwriters of North America, Artist Rights Alliance, Black Music Action Coalition Music industry brief in support of plaintiffs' summary judgment motions. The argument summary couldn't be pulled from the text. Open the PDF for details. ECF 2399-1 · Oct 30 Ashley Judd Draws on her experience as a source for the NYT's Weinstein reporting: work like that needs well-resourced human journalists and trusted relationships, which AI can't replace and threatens to defund. ECF 2402-2 · Oct 30 ASJA, ASMP, American Literary Translators Ass'n, National Writers Union, Novelists Inc., Romance Writers of America, Sisters in Crime, Dramatists Guild, SFWA, Nat'l Ass'n of Science Writers, NPPA, Garden Communications Supports the class plaintiffs. Downloading from LibGen and Books3, training, and OpenAI-Microsoft distribution are not fair use. AI output floods the market with substitutes, and defendants acted in bad faith. ECF 2404-1 · Oct 30 Reps. Jamie Raskin, Jerrold Nadler, Hank Johnson DOJ's national-security argument amounts to an unauthorized blanket exemption for all AI training. It deserves no deference, and national security isn't a fair use factor. Styled as a response to the DOJ Statement of Interest. ECF 2406-1 · Oct 30 Ken Burns Training on his films is destroying the market for quality historical documentaries, and the models regurgitate exact copies of his material. ECF 2413-1 · Oct 30 PEN America, Protect Democracy Project, Reporters Without Borders (RSF USA) Defendants' arguments invert the historical relationship between the Press Clause and the Progress Clause. Their business practices threaten a free, independent press. ECF 2417-1 · Oct 30 Cate Blanchett, Steven Soderbergh, Nick Hexum, CAA, Creators Coalition on AI, Music Artists Coalition, Nat'l Ass'n of Voice Actors, Human Consent Foundation, Writer, Inc. A working licensing market for training data exists. Collective licensing and machine-readable consent tools are in use, and that should drive the factor four analysis. No explicit ask; the argument favors plaintiffs. ECF 2420-2 · Oct 30 Copyright law professors (incl. Julia Powles, Trevor Reed, Marketa Trimble, Bhamati Viswanathan; full list in the brief's appendix) Training and RAG copying is direct infringement. In these works expression is inseparable from facts and ideas, and copyright has never allowed wholesale copying of expression to reach the underlying knowledge. ECF 2421-1 · Oct 30 Progressive Democrats of America, RadicalxChange, Tech Oversight Project, Young People's Alliance, Prof. Walter Scheirer (lead author David Atkinson) Defendants want "GenAI exceptionalism." If their use is fair, almost any human's unlicensed use would be too. Public availability doesn't create an implied license. ECF 2438-1 · Oct 30 Prof. Joseph Stiglitz Economic argument: unlicensed AI use erodes incentives and degrades the information ecosystem, and the split of value is grossly inequitable. Licenses should divide the gains. Attaches an economics paper (51 pp.). ECF 2448-1 · Oct 30 America's Newspapers Loss of local news makes government costlier and less accountable, communities less safe and democracy weaker. Claimed public benefits can't outweigh that. ECF 2452-1 · Oct 30 Jim Albrecht Focuses on RAG: the copying is the violation and the output shows a substitutive purpose. Harm to news is ongoing, and spreading the copying across sessions doesn't excuse it. ECF 2457-1 · Oct 30 AAN Publishers, Inst. for Rural Journalism, LION Publishers, Nat'l Ass'n of Hispanic Publications, Rebuild Local News Coalition, Solutions Journalism Network, NewsGuild-CWA, SPJ Strong copyright gives small outlets leverage for compensation. Today only big outlets get AI deals, so a ruling for the NYT would push AI companies toward licensing that pays local news. ECF 2460-1 · Oct 30 AI startups: BandLab, Boomy, Created by Humans, Credtent, Splice, Newstex, ProRata.ai, SourceAudio, Cashmere (Vandal AI), Voice-Swap Licensed training hasn't priced developers out. Licensing creates markets, pays creators and reduces legal risk, so requiring permission won't slow innovation. ECF 2099-2 · Oct 14 The AI Rights Project Defendants haven't shown that machine-scale training is equivalent to human creative reuse, which is the premise of transformativeness. Rebalancing the IP Clause is for Congress. The motion says it supports plaintiffs, but the brief "urges no outcome." ECF 2456-1 · Oct 30 William Morris Endeavor (WME) Factor four: AI companies already license copyrighted works, voices and likenesses, and WME negotiates those deals. Unauthorized training and AI replicas harm that market, and lost licensing fees count as market harm. Supports the news plaintiffs. Filed only in the NYT case (23-cv-11195). Neither party / mixed (4) ECF 2361-1 · Oct 29 Wikimedia Foundation Analyze each use separately under Warhol. Scraping itself isn't a "use." Pre-training is likely fair use; feeding copied works into a chatbot's RAG system likely isn't. ECF 2408-1 · Oct 30 Center for AI and Digital Policy, plus AI experts and legal scholars Training-data provenance is central to accountability. Identify each use separately, and don't let poor recordkeeping become an evidentiary advantage. Leans plaintiffs. ECF 2451-1 · Oct 30 Institute for Intellectual Property & Social Justice (Lateef Mtima) Asks the court to weigh social-justice values such as equitable access, inclusion and empowerment, beyond the usual economic framing. The specific ask is unclear from the excerpt. ECF 2434-1 · Oct 30 Michael Moritz LLMs and independent journalism both matter, and LLMs depend on continued quality reporting. For the two to coexist, a workable licensing market has to develop, and that will happen only if the law makes clear developers can't copy publishers' content without paying. Filed in the NYT case (23-cv-11195) and supports neither party, though its argument leans toward the news plaintiffs. Discloses that a co-author was formerly seconded as counsel to the Times. Brief not yet located (1) ECF 2432 · Oct 30 Profs. Jon Garon, Adam Mossoff, Loren Mulraine, Chris Newman, Sean Pager, Zvi Rosen, Mark Schultz Likely supports plaintiffs; this group usually files pro-rightsholder briefs. Only notices of appearance (2432, 2450) are in RECAP. As of Oct 31, no brief appears on the RECAP docket after 2462; check PACER.

Sides are assigned by what each brief says it supports; where a brief claims neutrality but clearly lands on one side, it is placed there with a note. Summaries are drawn from each brief's Summary of Argument, Introduction or Conclusion. RECAP's copy of the docket stops at ECF 2462 (Oct. 30, 2026).

This article is republished from The Conversation under a Creative Commons license. Read the original article.

With the U.S. midterms this week, about 1 in 5 voters has asked AI chatbots about elections and politics, a recent survey found.

But many voters may not know that governments around the world increasingly shape what chatbots can say. And because many tech companies apply the same policies worldwide, the answers American voters get may be shaped by governments at home and abroad.

I'm a researcher at a nonpartisan think tank at Vanderbilt University, where I study free speech policy. In two reports published in 2025 and 2026, my colleagues and I worked with global legal experts to examine how governments in 12 countries and regions regulate artificial intelligence. Our research shows that they often use existing rules, such as national security or anti-state propaganda laws, rather than new AI laws to control what AI models can say.

Censorship masked as safety

AI can cause real harm, from launching cyberattacks to generating sexualized images of children.

It's important that governments tackle these harms. But, in our assessment, some governments use legitimate concerns like these as a pretext to control political discussion, particularly chatbot responses about those in power.

In 2024, an Indian journalist asked Google's Gemini chatbot whether the policies of India's prime minister, Narendra Modi, were fascist. Gemini responded that some experts say yes. The government reacted by requiring official permission before "unreliable" AI models could be released. Although authorities later dropped this requirement, the case shows how terms like "under-testing" or "unreliable" can be used for political purposes.

India is not alone. China requires AI models to uphold "socialist core values."

And in Turkey, a court blocked posts by Grok, the chatbot built into the social platform X, that allegedly insulted President Recep Tayyip Erdoğan, founding father Mustafa Kemal Atatürk and religious values. Research by Turkish law scholar Yaman Akdeniz identifies the Turkish rulings as among the earliest documented cases of state censorship of chatbots.

Vietnam shows what such content controls can mean for dissidents, including those in exile. A Vietnamese court sentenced two activists, who live in Germany, to 17 years each in prison in December 2025 for spreading anti-state material. Prosecutors accused them of using AI to fake voices and images of Vietnamese leaders, illustrating how prosecutions can target political criticism.

Foreign rules, domestic impact

Some of these government measures only affect people residing in their countries. Other statutes cross borders.

The European Union's AI Act, adopted in 2024, requires any company offering AI tools in Europe, wherever it is based, to mark AI-generated content. Anthropic, for one, applies the marks globally because it has no reliable way to apply them by region. This is an example of the Brussels Effect, in which rules adopted in the EU end up applying elsewhere.

European rules also affect what chatbots say about politics.

In August 2026, the EU placed ChatGPT under the strictest tier of its Digital Services Act, Europe's online safety rulebook. OpenAI, creator of ChatGPT, now must assess and reduce the "systemic risks" of its service, including risks to elections, such as the spread of disinformation.

And the Dutch Data Protection Authority tested four chatbots before the 2025 national election and found that the political middle ground "virtually disappears" in their voting advice. It urged providers to "prevent their systems from being used for voting advice."

These rules can be used to pressure platforms.

In 2024, Thierry Breton, then the EU's top digital rules enforcer, reminded Elon Musk of his legal obligations before Musk's live interview with then-presidential candidate Donald Trump on X, warning against the spread of "harmful content."

The United States is no exception

Threats to Americans' access to information do not come only from abroad.

In 2025, President Trump issued an executive order on "Preventing Woke AI in the Federal Government." It requires the AI models the government buys to be "truth-seeking," ideologically neutral and not favoring ideas such as "diversity, equity and inclusion."

"Truth-seeking" may sound like an objective, neutral term, but it can mean very different things to different people. Take media reporting: The president recently banned three news outlets from the White House for reporting "FICTION and LIES" and dismissed media polls as "fake."

More recently, the Federal Trade Commission has suggested that AI companies that steer answers toward undisclosed ideological or biased objectives might break consumer protection law.

A 2025 Pew Research Center survey found that 8 in 10 Americans believe Republican and Democratic voters cannot agree on basic facts. In other words, as our report notes, government mandates to seek truth are often driven by ideological belief systems, particularly on political issues.

Meanwhile, dozens of states have passed laws restricting AI-generated political deepfakes. California and Hawaii, for example, are two states that have banned sharing deceptive AI content about candidates in the months around an election. U.S. courts have blocked both of those laws, citing the First Amendment.

What does this mean for voters? When you ask your chatbot about the elections, treat the answer as a starting point. Keep in mind that governments, both foreign and domestic, may try to influence the answer. AI responses should not replace your critical thinking at the ballot box.

Jordi Calvet-Bademunt is a Research Fellow and Visiting Scholar of Political Science at Vanderbilt University and The Future of Free Speech think tank.

The Ultimate Electric Circuits Engineer Course Bundle is designed for beginners and students wanting to build knowledge in electrical engineering and circuit analysis. You'll start by learning the fundamentals like Ohm's law and Kirchhoff's Laws before moving into advanced network theorems. You will learn how essential components like operational amplifiers, capacitors, and inductors behave, while also gaining hands-on experience with alternating current and MATLAB simulation tools. It's on sale for $30.

Note: The Techdirt Deals Store is powered and curated by StackSocial. A portion of all sales from Techdirt Deals helps support Techdirt. The products featured do not reflect endorsements by our editorial team.

By now it's likely you've heard of the whole concept of the AI paperclip maximizer problem — a hypothetical first put forth by Nick Bostrom — suggesting that a misaligned AI system that looks to optimize on the wrong single thing (in this case creating paperclips) could lead to the end of humanity. Specifically, the thought experiment is that if you have a sufficiently powerful AI tool that is told to "maximize" the creation of paperclips, it might decide the best way to do that is to seize control of all resources and dedicate them solely to making paperclips. And, naturally, that might include grinding up humans, either as raw material for paperclips or as fuel to power the paperclip factory.

The supposed "lesson" of this is to be cautious about these AI tools, though I think the actual lessons are twofold: (1) thought experiments can only take you so far and (2) never over optimize on a single metric. Indeed, this was part of what our recent series, The Metric is Not the Mission, was all about.

Anyway, everyone in the AI world knows about the whole paperclip thing. I guarantee you Sam Altman knows about it. He has praised the book where Bostrom put forth the hypothetical, claiming that it was "the best thing I've seen" on the topic of "will AI kill us all." And he's joked about it in a memey way:

But it's not clear that he truly has internalized that lesson regarding not optimizing on just a single thing. Last week he went on John Collison's (co-founder of Stripe) "Cheeky Pint" podcast to talk about the state of AI.

A bunch of the headlines focused on one specific thing that he said, however, that it's "quite terrible" for people to be wasting RAM in home machines over having them working full steam in a data center somewhere. The context here matters a bit. Collison catches Altman a bit off guard by asking "how's the compute shortage going?" just as Altman takes a big gulp from his pint of beer, and you see Altman try not to laugh/spit out his beer, before there's an obvious cut in the video, where he recomposes himself.

They then have a discussion about the increasing demand for compute, Altman talks about how they're working on having robots build "self-replicating data centers" (what?!?) to which Collison rightly interjects that the problem with data centers right now isn't the lack of labor available to build them. Collison then shifts the conversation back to the stuff that most people care about: the fact that RAM prices have increased dramatically recently, which is jacking up the price of everything from computers to phones.

Collison notes that people keep saying the market should take care of this: if prices keep going up, shouldn't Adam Smith's "invisible hand" motivate more companies to make more RAM and ease the shortage? Altman shows that he's not a big believer in the invisible hand of the market concept:

Collison: My observation, the compute stuff is they say: "The solution to high prices is high prices." And it's like, okay.
Step 1: We have the high prices for memory, and GPUs, and everything. The solutions should be here any moment now. How are we doing on the solutions front?

Altman: Bad.

Collison: And I'm not saying this is OpenAI's job. I feel this is the invisible hand's job, right? Memory's gotten really expensive. So, new memory makers are going to flood the market any moment now.

Altman: I think the invisible hand is failing right now. I mean, the invisible hand… It has like a checkered track record at best. But, I think it's not quite working here.

He then points out that, for decades, DRAM was a commodity product with razor-thin margins and barely any profits. His argument is that memory makers aren't willing to expand much because they're afraid that the bubble might burst and they'd have overbuilt (which… might not be wrong?).

Of course, Altman leaves out of this the fact that OpenAI has basically tried to corner the market on RAM, trying to lock up 40% of the world's supply without even guaranteeing that it will follow through on the actual purchase orders. I mean, I think that alone would have at least some impact on the current price for RAM, wouldn't it?

Collison pushes a bit further, and still wants to know why others aren't getting into the memory market, and even (half jokingly?) suggests Intel could get into the market — a reference to one of the key parts of Intel's lore and mystique: the Andy Grove decision in 1985 to get completely out of the memory chip business to focus on processors, which got a bunch of recent attention when Intel's current CEO suggested the (struggling) company might reverse that decision 40 years later.

That brings Altman around to the critical quote regarding local RAM, in which he dismisses the argument that local inference is important, as a way of suggesting that the more efficient use of memory would be in large datacenters where it can be utilized at a much higher rate:

Oh, I think there are all of these ways where the supply chain—it's just super non-obvious.

An example of this is people…A lot of people want local inference. And somehow, they assume that the problem is solved now. Because, we are going to have open-source models. We're going to be able to run them on our computers. And so we're not going to need all the stuff in the data center.

But from a global allocation perspective, actually, it's quite terrible to have everybody have a bunch of the world's limited supply of DRAM sitting in their own computer. And, being used—Mm-hmm. 5% of the time. You want maximum utilization in the cloud if you're trying to deal with this very supply constrained environment.

And somehow everyone's… many people's intuition is the opposite of that.

So he's correct… in one very, very, very narrow sense of this quote. If you are only interested in dealing with the current "world's limited supply of DRAM" he's not wrong that it is more "efficient" to have that RAM be in a data center and then rented out to you. And he's not exactly wrong that if everyone just paid per token for remote compute, the cost of RAM in your own PC would matter a lot less.

But… he's kinda missing the larger point: that people are not just focused on optimizing for the most efficient use of memory. They're not just optimizing for what's best for "the global supply chain of RAM makers." They have other concerns — and other uses beyond AI — as well, such as their own sovereignty and control over their own tools and systems. There's a reason my essay on the need for decentralization (including in AI) went so viral a few months back.

And it's not because people think they're single-handedly going to solve the global supply crisis in DRAM by running everything locally. It's because we've seen what happens when a central provider controls key parts of our stack, and the results are very, very, very bad.

Using open weight models on local hardware may not be the most efficient for the exact price of RAM (which, again, Altman tried to corner the market on), but it certainly leads to much more actual user empowerment by people who want to use these tools, but without outsourcing control to someone like Altman.

In a pure "efficiency" model where you are solely optimizing on what's the most cost efficient way to use memory the answer would be to have it in datacenters where you'd benefit from both economies of scale as well as flexible utilization. But that has consequences. It's the same logic by which a fascist dictator may seem "more efficient" to those sick of democratic quirks and bureaucratic red tape. It only sounds good if your mental model is completely screwed up and simplistic.

In both cases, you're talking about a very simplified mental model, optimizing on a suboptimal point. Personal autonomy over our tools and our systems is more important than making your data center more efficient, Sam. The paperclip problem comes from focusing on a narrow metric and over optimizing on it. And in this clip, Sam is paperclipping RAM.

Recently, reports emerged that LG TVs are sniffing around your local network, and even listening to your local environment, even when the TV is purportedly "off." Some of those voice commands were then stored in plain text and uploaded to the internet. LG's response? To deny everything.

For decades now we've documented how while "smart" TV makers are obsessed with making money off of in-home surveillance (which is now more profitable than the actual TV), they do an abysmal job protecting the data they collect. They also routinely don't respect consumer choice when users say they don't want to participate in the full scope of hyper-commercialized mass surveillance.

That latter point is central to a new study out of Northeastern that conducted 5,943 experiments across twelve commercial "smart" TV devices. The researchers found that most smart TVs utilize automatic content recognition (ACR), a technology that captures audio or video snapshots of viewed content and then monetizes that data to help build personalized profiles of viewers.

TV makers insist that users are supposed to be able to opt out, but the choice simply doesn't work across a large segment of TVs. Companies like Vizio simply ignored the consumer requests entirely more often than not:

"We find that the advertised ACR opt-out fully suppresses ACR-attributable traffic on only four operating systems; on Samsung Tizen traffic consistent with ACR activity persists in more
than 60% of evaluated scenarios, and on Vizio SmartCast a content correlated endpoint hosted on Amazon Kinesis continues to receive traffic in 100% of opted-out scenarios."

These companies have been awful for a long while now, and U.S. regulation of the sector has varied between soggy and inconsistent. Many smart TV manufacturers fail to encrypt sensitive data, or remove features if you refuse to have your daily viewing habits measured and monetized (by say, disconnecting your TV from the internet).

Vizio was fined $2.2 million by the FTC back in 2017 for failing to tell customers that their televisions were spying on them. The company also wound up facing a $17 million lawsuit that wound up giving consumers somewhere between $13 and $31 (lawyers in the case saw $5.6 million in fees).

The researchers say since they can't see the precise scope of what's being collected due to encryption, they're unsure whether Vizio is technically violating its FTC agreement (not that we have a current FTC that would care if they were):

"The opt-out setting has no observable effect on traffic sent to this endpoint; however, we cannot inspect the payload due to encryption so we cannot confirm whether this violates the FTC settlement."

Results of whether opting out of ACR tracking actually stopped ACR tracking were all over the map. Roku, for example, ignores opt-out requests when users are watching Roku's own channels. None of the smart TV vendors track Netflix viewing for fear of violating agreements with the streaming giant. Samsung's opt out function was more likely to actually work when a user was watching antenna broadcasts.

On two Smart TV platforms, personal viewing data collected by ACR remains unencrypted, despite years of bad press and scandal on the subject. Many appear to be violating the law, but because U.S. regulatory oversight has been gutted, they see no meaningful accountability. That includes smart TV OS' like VIDAA, which is most prominently found on Hisense and Toshiba smart TVs:

"VIDAA OS sends ACR data to servers run by Unruly over MQTT without encryption, so anyone on the network path between the TV and those servers can read it. Each message carries a Client ID, a permanent identifier for the TV that can't be reset, along with the time zone, the app name, the input (antenna, HDMI), the antenna channel and an audio fingerprint.

The app name alone can be revealing. We saw ACR on apps such as VIDAA Kids, which raises concerns under the U.S. children's privacy law (COPPA), and Christianity and Money. VIDAA's own HOME OS app sent show and film titles in place of the app name."

Keep in mind these researchers just focused on ACR — or video, image, or audio snippets taken to track your viewing and build a profile on you. There are numerous other ways smart TVs can track you — and numerous other avenues via which your traffic is being surveilled by everybody along the chain, including your telecom provider.

This persists because companies do not fear U.S. regulators, because U.S. regulators have been largely lobotomized via corruption. The United States also remains too corrupt to base even a baseline privacy law for the internet era or regulate the data brokers that further monetize much of this data. Still the researchers were nice enough to provide regulators a path, should they someday choose to do their jobs.

This week, our first place winner on the insightful side is the first of several comments about Trump's voting fraud squad proudly announcing the arrest of… 28 people in California. It's an anonymous comment summing up their attitude:

Republicans just want election fraud. That's it.

In second place, it's a double-winning comment that also took second place over on the funny side. Once again it's an anonymous comment, this time about Ofcom ramping up the Online Safety Act after all the evidence that it isn't working:

In other words, the Online Safety Act has one, and only one, genuinely good provision: it requires Ofcom to report the data showing that it doesn't work.

For editor's choice on the insightful side, we start out with Stephen T. Stone and one more comment about Ofcom, this time answering the question about why they're doing all this:

Simplest answer is often the best: They want to use this as a pretext for an Internet censorship endeavour in the vein of China's Great Firewall.

Next, it's Rocky with another comment about the 28 voter fraud arrests:

Which also means if they can only find 28 erroneously registered people out 23 million it is proof that the system works because no system is perfect and a error rate of 1.22e-6 is so incredibly low it is statistically insignificant. I expected it to be much much higher, because as you say, voters screw up and voters are people and people screw up daily.

I'm just waiting for the idiots to show up ranting about "see, I told you!" which means they think they are perfect and have never ever in their life made a mistake which is utter bullshit. But that's they way of the fools, confidently ignorant of their own shortcomings.

Over on the funny side, our first place winner is David with a reply to another comment aptly describing the rate of voter fraud as "one percent of one percent of one percent":

That's already three percent, enough to swing an election.

And don't tell me that isn't how percentages work, or you won't get to enjoy 2000% percent reductions in drug prices.

We've already had the second-place double-winner above, so we'll move on to editor's choice, starting out with one more anonymous comment about the fraud arrests. This time, it's about the government's little trick of lumping together "illegal registration or voting" as a single stat:

Hundreds of millions of Americans have committed murder or drank water.

Finally, we've got an anonymous comment about RFK Jr. personally stepping in to approve peptide use over the FDA's objections:

My peptides bring all the Übermenschen to the yard,
And they're like "OH MY GOD WHERE DID ALL THESE TUMORS COME FROM?"

That's all for this week, folks!

This Week in 2016

• Dumb & Dumber Claims About Last Week's Internet Attack (SOPA?!? Really?)
• Shake Up At The Copyright Office A Possible Preview To Fight Over Copyright Reform
• The Reason The Copyright Office Misrepresented Copyright Law To The FCC: Hollywood Told It To
• Facebook's Arbitrary Offensiveness Police Take Down Informational Video About Breast Cancer Screening
• As Expected, FCC Passes Modest Privacy Rules For Broadband Providers, ISPs Act Like World Has Ended
• MPAA: EFF Just Jealous It Doesn't Control Copyright Office Like Hollywood Does

This Week in 2011

• Universal Uses Copyright To Censor Bad Lip Reading Parody; Why Not Embrace It?
• The Many Killers Of The Film Industry: Volume One
• The Many Killers Of The Film Industry: Volume 2 - A Disaster Called Television
• The Many Killers Of The Film Industry Vol. 3: Post-Ellipsis And Beyond Television
• House Trying To Rush Through Its Version Of PROTECT IP; Tech Industry Asks Why?
• PROTECT IP Renamed E-PARASITES Act; Would Create The Great Firewall Of America

This Week in 2006

• Copy Protected Music Ghettos Are Holding Back The Industry
• Network Neutrality Debate: It's All Relative
• Newspaper Pretends It Can Block Fair Use And Threaten Those Who Say It Can't
• MPAA Helps New York City Down The Path Of Pointless Legislation
• Using Trademark Law To Stop Free Speech
• RIAA Still Not Allowed To Get Free Access To Scour Your Hard Drive

If you go through all of the bloviations from RFK Jr. when it comes to his stance on vaccines, one of his chief refrains is that vaccines that have been in wide use for many years haven't been tested properly for safety, or that data about their safety has been hidden or ignored. He's generally wrong about all of that, of course, but it doesn't stop him from saying it over and over again. The end result, if you didn't bother looking any further into what Kennedy has said and written on the topic, is that you might believe that what Kennedy really cares about is a cautious approach with stringent testing before allowing something to be put into American bodies.

But that isn't what Kennedy believes. It's not what he demonstrates through actions, either. And if you want proof of that, you need only read about how Kennedy has unilaterally adjusted rules at HHS to allow pharmacies to sell you peptides over the objections of the FDA.

Anti-vaccine Health Secretary Robert F. Kennedy has unilaterally passed a rule that will allow specialty pharmacies to sell six specific peptide drugs, overruling scientists at the Food and Drug Administration.

Kennedy's new rule sidesteps the need to formally undo that FDA action. The rule is titled "Enforcement Discretion Policy Regarding 503A Compounding Using Certain Peptide Bulk Drug Substances" and was made available Wednesday on the Federal Register, with the final publication date set for October 30. The rule does not change the regulatory status of the peptide; instead, it states the Department of Health and Human Services (HHS) will no longer take enforcement action against compounding pharmacies for making the peptides, essentially giving them a green light to freely produce and sell them.

So what qualifications does Kennedy possess that is informing his decision to overrule actual FDA scientists on the topic of peptides? None, of course. Kennedy is not a scientist. He's not a healthcare professional. What he is, instead, is both a user of peptides personally and someone who has many friends in the peptide business. It's just another grift, though it is being wrapped in pretend backing from the panels that Kennedy personally stocked with sycophants.

One reason cited for all the peptides was that an FDA advisory panel voted in July to add these six peptides back to the list of drugs compound pharmacies can make. However, prior to the vote, Kennedy stacked that 12-member committee with eight new members. At least six of Kennedy's additions have ties to peptide-related businesses and clinics, raising serious conflict-of-interest concerns.

Another explanation for allowing the peptides is that the safety concerns are mainly hypothetical or "inferential" and not based on observed harm in humans. But this is mainly because there are few to no studies involving humans that show either benefits or harms from the peptides.

And here we put the lie to Kennedy's concerns about vaccines. If he wants to claim that vaccines haven't been tested enough to be in wide public use, he cannot create a non-enforcement policy over peptides that have had far less, if any, such testing. We are living under health policies that are purely Kennedy's personal whims. That's nothing like the gold standard science that he has been promising to follow throughout his tenure at HHS.

The peptide pushers are celebrating and it's not hard to understand why. Many Kennedy allies are among them. He gave them a gift, almost perfectly circumventing his previous fears about testing to do so.

There's a lot wrong with AI. But bad things happen when we try to use copyright law to fix it. Which is what we just told the Southern District of New York, where OpenAI and Microsoft are defending themselves in several cases alleging that their training violated the copyrights of the works the AI models were trained on. We made a few points.

One overarching point is that what the plaintiffs are trying to do is not what copyright law is for. Copyright gives the holder the ability to control how the embodiment of the work is used (ex: you can't copy a book without permission, or, as the example we used in the brief, perform a play) but not the ability to control the public from benefiting from its contents. Copyright can control such activities as copying, distributing, or performing a work but NOT reading, watching, listening to, or learning from a work. What OpenAI does with its training is the latter set of activities, not the former. But if OpenAI is somehow not able to engage in these activities because of copyright, then no one can. Which would lead to a direct harm to the public, a loss of an important benefit copyright law is actually supposed to give it.

In fact, it is a benefit copyright law must provide the public. The whole point of copyright law is to promote "the progress of sciences and the useful arts," or, in other words, help ensure we get more innovation and expression. But the plaintiffs in these cases want the opposite: they want copyright to stop these things. They want copyright law to prevent AI from being further developed (indeed, in its complaint the New York Times Company even asked for OpenAI's models to be destroyed!) by denying it the ability to do the learning that is needed for it to be useful. And they want to prevent any of the further expression that AI helps facilitate, even though fostering new expression is exactly what copyright law is supposed to do.

In fact, it is what copyright law is constitutionally required to do. Moreover, if, rather than fostering innovation and expression, copyright law instead obstructs it, then copyright law is not constitutional. Congress only gets its authority to make copyright law through the Progress Clause ("to promote the progress of science and the useful arts…") so if it turns out that the law it has passed instead does the opposite of promoting progress, then that law is unconstitutional, as being beyond its authority.

The First Amendment also limits Congress's authority to pass laws that inhibit free expression. So Congress also would not have had the authority to pass a law that let copyright holders of existing works prevent either the development of AI or any of the downstream, lawful expression AI helps facilitate. If the copyright statute Congress passed were to have that effect, then it would be unconstitutional.

It also would violate the Constitution in another way we've discussed before: by violating the First Amendment-protected "right to read," which is sometimes also referred to as the right to receive information. All those learning activities (reading, watching, listening to, etc.), which the current statute does not actually restrict, are activities that the First Amendment entitles the public to do, and copyright law not only does not prevent, but constitutionally could not prevent.

Meanwhile, we also made an overarching point: even to the extent that AI as we know it may be terrible, and we are looking for a law that can stop it, copyright law is still not the right law to do it. First of all, it's not tailored to the actual problems we'd want to address (for instance, if using AI in hiring decisions produces discriminatory results because of bias incorporated into the AI model, then the solution to that problem is making sure that AI can train on even more material, not less, because it is the training limits that tend to create that bias). Secondly, the worse we think today's AI is, the more important it is that something better is developed, but if law can stand in the way of further development, then none will be. None can be, and we'll be stuck with the technology we have, if not an even worse version of it if we have to degrade the models in response to copyright claims. As well as stuck with an imbalanced, super-powered copyright law that could stand in the way of lots of other non-AI innovation and expression.

And, again, if we have to freeze the technology because of copyright law, then copyright law is definitely running afoul of the Constitution, which allows Congress to pass a copyright law that only does the opposite.

In sum, the worse you think AI is—and we make no quibble with that assessment, although we did point out that it also has some distinct upsides, particularly when it comes to enabling more expression—the more important it is that these copyright claims be dismissed. Because however bad the state of play may be today, a copyright law that would let these claims prevail will only make everything worse.

Ctrl-Alt-Speech is a weekly podcast about the latest news in online speech, from Mike Masnick and Everything in Moderation's Ben Whitelaw.

Subscribe now on Apple Podcasts, Overcast, Spotify, Pocket Casts, YouTube, or your podcast app of choice — or go straight to the RSS feed. To get extended episodes with additional coverage, support us on Patreon.

In this week's episode, Mike and Ben cover:

• Lula's Win Signals Fight to Regulate Social Media Will Escalate in Brazil (TechPolicy.Press)
• Meta Employees Denounce Interference in Brazilian Elections (Panamerican Dispatch)
• Google's AI Chatbot Refuses to Answer Question of Who Won 2020 Presidential Election (Mediaite)
• AI-driven disinformation running rampant on social media before US midterms, experts warn (The Guardian)
• Never Mind The Deepfakes, Chatbots Have To Deal With Election Lies Being Free, While Truth Is Paywalled (Techdirt)
• French Government Calls Emergency Meetings With Platforms Over Protests (Wired)
• Farewell, France (Quad9)

And in the extended episode for Patreon supporters, they cover:

• Ofcom's Fix For An Online Safety Act That Isn't Working: More Online Safety Act (Techdirt)
• How age checks at age 16 can work to underpin UK's social media ban (Ofcom)

Our fun links this week include Krillion (much harder than it looks) and Dial Flipper for discovering new podcasts.

Follow us on Instagram, YouTube, and Bluesky for video clips from the episode!

If you're already a Patreon supporter, you can get the extended episode on Patreon.

We have talked about the Wyden Siren many times before. Senator Ron Wyden learns, through classified briefings, that the intelligence community is doing something most Americans would consider a clear violation of their civil liberties — often justified by interpreting plain English words to mean something no other English speaker would recognize. But because that information is classified, he can't just say what it is. So instead he sends a letter demanding that the intelligence community explain itself, carefully worded to tip off the public that the government is effectively lying about how it's violating our civil liberties. He's done it many times, from warning that the government had secretly reinterpreted the PATRIOT Act to flagging that supposed "cybersecurity" legislation was really being pushed for a more nefarious purpose.

This year alone, he's issued Wyden Siren letters a few different times. And now he's done so again, once again about Section 702 of the FISA Act, which has been a Wyden target for years. Wired has a summary:

Two Democratic senators who have long overseen US intelligence say the Trump administration is illegally withholding a secret court's findings from Congress until lawmakers renew a program that lets federal agents read Americans' communications without a warrant.

Seven months ago, the Foreign Intelligence Surveillance Court objected to search filters that let intelligence analysts pull Americans' messages from intercepted data without the approvals, written justifications, or records the law requires. The court approved another year of surveillance anyway. The ruling has never been made public, and Senators Ron Wyden of Oregon and Adam Schiff of California say the administration's own deadline for releasing it passed in September.

The program, Section 702 of the Foreign Intelligence Surveillance Act (FISA), lapsed in June after Congress failed to renew it, though collection continues under the court's March approval. The administration's own public documentation says rulings like this one are reviewed for public release "within 180 days of issuance." That window closed September 13, even as intelligence officials press lawmakers to reauthorize the program by the end of the year.

Wyden sounding the alarm is expected (and welcome). Schiff co-signing it is, however, a surprise. For many years when he was a Representative (prior to recently moving over to the Senate), Schiff was a reliable voice and vote for greater surveillance powers. He would frequently roll out his "as a former prosecutor" line to defend giving more and more power to the government.

Indeed, back in 2018, we called out Schiff (who was then the House ranking member on the Intel Committee) for giving Donald Trump greater surveillance powers in renewing Section 702. It was bizarre: Schiff was one of the loudest voices calling out Trump's abuses of power during his first term, and yet he led the charge among Democrats to hand Trump even more surveillance power. He even undercut Wyden's own attempts to rein in those powers, as Trevor Timm noted at the time in an op-ed blasting Democrats for handing Trump more surveillance power:

Schiff was a leading driver in the House to extend the NSA's surveillance powers, and has been undercutting the more robust reforms proposed by other Democrats, like longtime Senate Intelligence Committee member Sen. Ron Wyden, for months.

Given that history, seeing Schiff co-sign a letter accusing the administration of illegally hiding a ruling about 702 abuses is quite the turnaround.

Indeed, Schiff even blasted out a press release talking about "serious abuses of American's rights." That's quite a statement given how strongly Schiff fought to give those powers to the government.

The specifics here involve a secretive FISC ruling regarding Section 702. Technically 702 expired back in June, after Trump tried to tie its renewal to his ridiculous voting bill that no one is willing to pass. But back in March there was a secret ruling by the FISA Court which allows the government to keep using this surveillance authority at least until March of next year. That same ruling also objected to the search filters analysts use to pull Americans' messages without the approvals and records the law requires — and then approved another year of surveillance anyway. Wyden and Schiff want it public, and the law is on their side: the statute requires either a declassified version within 180 days or if the DNI waives declassification by yelling "national security" then it requires an unclassified statement from the Attorney General summarizing the legal interpretation. And Attorney General Todd Blanche unequivocally promised the Senate Judiciary Committee that he would deliver the unclassified opinion by the deadline.

That deadline came and went on September 13. The ruling remains classified. From Wyden and Schiff's letter:

We and others have repeatedly called for the administration to release the March FISC opinion. In May, the Chair and Vice Chair of the Senate Intelligence Committee sent a letter to the administration stating their expectation that the opinion would be released by May 15 "in order to inform the Senate debate on section 702." The statute itself requires the administration to release a declassified version of the opinion within 180 days. In July, Attorney General Blanche told the Senate Judiciary Committee he expected that the public release would happen "in advance of the statutory deadline." Attorney General Blanche unequivocally committed to "declassifying the opinion by the statutory deadline of 180 days and complying with the law." Nevertheless, it is now a month past the statutory deadline and the opinion has still not been released.

Simply put, this is not acceptable, especially as this administration restarts its attempts to lobby Congress to renew Section 702 without reforms. The administration has now had over seven months to declassify and release this opinion. It is time to stop violating the law and finally release this opinion. And if this administration won't follow this simple statutory directive, it is unclear why Congress or the American people should trust this administration to follow any of the other legal limits to Section 702.

We demand that you release this opinion as soon as possible. We also request that you respond in writing by November 6, 2026, providing an explanation as to why the opinion was not released by the statutory deadline and identifying the officials that were responsible for withholding its timely release.

As Wired notes, Senators can read the ruling, but only in a secure room, and those who have done so are dropping heavy hints about what it reveals:

Lawmakers who have read the ruling hint that it says more. Senator Dick Durbin, the top Democrat on the Judiciary Committee, urged colleagues on the Senate floor in June to read it in a secure room. "I have seen it," he said. Once they did, he told them, "your opinion of the role of FISA and the rights of Americans will be impacted for sure."

Wyden has also warned about a classified interpretation of Section 702 that he says multiple administrations have relied on and refused to declassify. "When it is eventually declassified," he said in March, "the American people are going to be stunned that it took so long and that Congress had been debating this authority with insufficient information."

Of course, the Trump administration seems to believe that laws are meaningless, especially if they require transparency or limit the government's ability to abuse the rights of anyone in America. So don't hold your breath waiting for Director of National Intelligence Jay Clayton to cough up this ruling. But as Wyden and Schiff point out, if the administration can't be bothered to follow a simple disclosure deadline, why should anyone believe it's respecting any of 702's other limits while it begs Congress for a clean renewal?

The best way to learn Raspberry Pi is by doing and in this course, you'll get familiar with specific features by building your own GPS tracking system. From setup and configuration to location tracking through Google Maps, this course will help you go from beginner to advanced with Raspberry Pi. It's on sale for $50.

Note: The Techdirt Deals Store is powered and curated by StackSocial. A portion of all sales from Techdirt Deals helps support Techdirt. The products featured do not reflect endorsements by our editorial team.

Well well well. All it took was a regime change and the persistent abuse of camera networks by police officers to change MapQuest from "Remember MapQuest?" nostalgia to "hey where can I download MapQuest" demands.

First, there's the regime change. Trump has been stupid as fuck all the time since day one but nothing was much stupider than issuing an executive order renaming the Gulf of Mexico as the "Gulf of America." Oh wait. There was more stupidity. Trump — again saying anything that momentarily glanced off the shiny, wrinkless lobes of his forebrain — demanded Lake Ontario be renamed "Lake America."

If that were the end of the story — a wet-brained spray-tan-maxxing fuckmook participating in some short attention span jabbering between hours of Truth Social doomscrolling — we wouldn't be here today discussing this latest turn of events.

But Trump said something and companies started doing something. Some even started doing something before the spittle dried on Trump's lips, as Karl Bode discussed earlier:

Google was the first to quickly make the change to appease Trump; in fact they acted so quickly on this the change to Google Maps happened before the government had even finished implementing it. Apple was very quick to follow suit, despite the fact that nobody at the company actually supports the ridiculous and pointless change.

That's how tech billionaires greeted this momentary outburst from a guy who's never bothered to think twice. One company altered its map before it was even made official. The other did it even though no one in the company actually wanted to have it done. These are the actions of two of the richest and most powerful companies in the world — companies whose fortunes would have dipped to the equivalent of a rounding error if they'd chosen to reject this overture.

Meanwhile, one of the first names to the (map) game refused to play along. MapQuest doesn't have billions of revenue to spare, much less the ubiquity of companies that offer phones as well as pre-installed map apps. MapQuest once owned the digital mapping world, but it has been sidelined by vertical integration.

Maybe it's just because MapQuest (relatively) has nothing to lose. Or maybe it's because MapQuest actually seems to care about its users. Either way, MapQuest refused to alter its map to reflect Trump's digital takeover of Lake Ontario.

And now MapQuest is going after America's surveillance state — one made most infamous by another company that has far more liquidity than MapQuest: Flock Safety. While it's certainly not the first company to offer automatic license plate readers, it's the most famous. Its network is huge, capable of capturing 20 billion plate reads a month. And it's the current favorite of law enforcement agencies all over the US.

But Flock is working against its own ubiquity at the moment. It's more famous now for allowing cops to stalk people, especially ex-wives and girlfriends, as well as any people their exes might now be seeing. It's also a handy tool for cops who just want to track people, including anti-police violence protesters or anyone else who's annoyed them.

People want to know where these cameras are. Flock's tools have generated a ton of public records that have ultimately given activists the ability to locate and map camera locations. Other companies working with law enforcement have generated some data as well, albeit not as much as the current market leader.

MapQuest now offers what other map apps don't: a handy map of ALPRs.

MapQuest users in Canada and the U.S. can now see where some surveillance cameras have been set up in their communities.

The company introduced a "surveillance layer" earlier this week that uses animated eyes to mark where automated licence plate readers, or ALPRs, have been set up by police or others.

"We knew that there were a ton of cameras in use and documented in the U.S. and Canada," MapQuest general manager Doug Berger told CBC News. "You would never know unless they were pointed out to you."

This is very cool. For multiple reasons. A company deciding to highlight law enforcement surveillance on its official app is completely unexpected. That it did this because some of its users asked it to is even better, which means MapQuest cares more about its (relatively) small number of users than Google and Apple care about their millions of users who default to their apps simply because its part of their smartphone's pre-installed ecosystem.

And that means MapQuest will continue to grow, three decades after its debut. GM Berger noted that the company gained three million new users after refusing to rename Lake Ontario. Adding a "surveillance layer" just means more deliberate interaction from users. And people who choose apps for a reason, rather than just using whatever first presents itself to them, are far more loyal.

This is a very smart move by MapQuest. The decisions it has made in the last couple of months shows it's far more respectful of its current and future users than the massive tech companies that assume (often correctly) that consumers will eat whatever shit is shoveled at them by their service providers. MapQuest is looking for people who want to use MapQuest because they like what the company is doing. It's an incredibly smart move that builds the sort of trust and loyalty larger companies can only wish they had.

Ideally, I think it makes sense to open up the U.S. market to Chinese goods (AI, EVs, robots), but have functional and well-funded regulators that police all labor, competition, environmental, privacy, national security, and consumer rights abuses. But big U.S. companies don't much like that because it would boost competition and regulatory accountability, eroding quarterly revenues.

So instead we get corruption-fueled protectionism run by a bunch of racist, and incompetent clowns who gutted most of our federal regulators, then saddled them with an elaborate new policy ask that's impossible to implement, even if the admin were competent.

Which it isn't.

For example, I recommend this recent Verge piece about the Trump effort to ban Chinese drones, which notes how it has been trivial for Chinese companies to trick the administration and bypass the ban simply by renting a strip mall office, tweaking a few names, and pretending their equipment is produced domestically. Curiously, officials who don't believe in government can't govern. Fancy that.

Another report this week by Ars Technica indicates that U.S. chip giant Intel has also been so disoriented and confused they felt it necessary to get a waiver from the Trump administration's "foreign robot ban" because executives didn't really understand what is or isn't covered:

"Intel's waiver request stated that it was "unclear" whether its robotic vehicles even qualify as robots covered by the FCC ban but said it sought relief just in case. The tech company also received approval to "make certain hardware, software, and firmware changes" that would help ensure the continued operation and maintenance of the robotic vehicles."

So it's a foreign robot ban, unless we need foreign robots. In which case big U.S. companies have to spend a bunch of time and money kissing the Trump administration's ass to ensure they have what they need to function. It's far from the "free market" dogma the administration rose to power on. It's a weird and racist autocratic nanny state operated by the dimmest people imaginable.

A recent study by the International Data Corporation predicts that the foreign robot ban could reduce US humanoid robot revenues by 58 percent over expected baseline. China is still expected to dominate the market with most of its cheaper, better gear avoiding the U.S. for the EU. And much like the FCC "drone ban," a big impact will be less tech product availability and higher prices overall for U.S. consumers.

Not content with screwing up the drone, robot, vacuum, and lawnmower sectors, the Trump FCC last week announced it would also soon implement a ban on "foreign surveillance chips." You're to ignore that the FCC can't really define what this means. And doesn't really want to answer any serious questions about how this will be implemented in a country already dominated by Chinese tech.

The Trump administration makes an endless stink about protecting U.S. cybersecurity defenses by taking aim at Chinese-made tech. But with its other hand it has devastated U.S. cybersecurity agencies with policies that are indistinguishable from a foreign attack, make it easier for foreign governments to surveil U.S. citizens at scale (the refusal to pass real privacy laws or regulate data brokers, for example), and function largely as a way for the administration to line its own pockets.

Again, protectionism is bad. But what we have here isn't really even that. There's zero real indication that the Trump administration has any idea what it's doing. There's simply no reality in which the Trump administration can implement or enforce any of this. And because our consolidated corporate press sucks, most coverage simply refuses to make this clear to anybody.

Outlets like Reuters, for example, inform the public that the Trump administration is "aiming to safeguard the U.S. artificial intelligence supply chain from Chinese threats." Others will parrot claims that the Trump administration is "dedicated to creating safety guardrails for AI." But there's no real evidence that's true or that the administration is capable of doing literally any of that.

This sort of pseudo-journalism will be all the rage in the new year as the Trump administration sets its sights on a likely ban of foreign AI models (to lock in the dominance of U.S. giants that dutifully kiss the authoritarian ring). Which is to say: expect this sort of normalization and sanewashing to increase despite zero evidence that any of this is working or makes any coherent sense.

To state that RFK Jr. says things that are just plainly wrong is to state the obvious. His waffling claims about the dangers of vaccines are wrong. His various other pet conspiracy theories, such as chemtrails and damage to humans from EMF, are also wrong. The data HHS is putting out under his leadership about how many measles deaths have occurred this year is also wrong. Hell, at a recent campaign event in Michigan, a state with a large Arab population, Kennedy decided that the best way to stump for Mike Rogers was to claim that none of the Arabs throughout the world produce any science or literature, which is also wrong.

"There's 500 million Arabs in the world, and yet they don't produce science, they don't win the Nobel Prize, they don't win the Fields prize, they don't win the Wolf Prize, they don't produce literature," he said.

Arabs have won all those prizes. Peter Hotez, who is a professor at the Baylor College of Medicine and specializes in vaccine development and tropical medicine, said his late colleague, Nobel Prize winner Ahmed Zewail, who was Arab and died in 2016, "would disagree" with Kennedy's comments.

So, not just wrong, but racist in perhaps the stupidest manner possible given his only reason for being in Michigan in the first place.

So let's rewind the clock back to April, when Kennedy decided to tell the public that this whole measles outbreak thing had nothing to do with him. Rather, the only reason we had the outbreaks to begin with was because of dirty foreigners coming in as immigrants.

But despite Kennedy being the most vocal source of vaccine misinformation, the secretary tried to blame the outbreaks entirely on immigrants who come to the U.S. from countries where measles is not eliminated — framing the issue as a global epidemic rather than a national public health crisis.

"It has nothing to do with me," he told lawmakers. "If you're worried about polio and tuberculosis, you should look at the immigration policies in this country. 'Cause the place where it's occurring are the place[s] where the immigrants are going, because they're not vaccinated."

That was wrong too, of course. We knew it at the time, just was we knew that the practice of blaming immigrants for disease outbreaks is such common practice that we could set our watches to it. But now we have actual data on how the measles is spreading in America and that data fully puts the lie to Kennedy's bigotry.

Most of the meat of the study focuses on very specific instances in which Washington state endured measles infections that were spread specifically from South Carolina earlier this year.

The study in Washington relied heavily on epidemiological data, linking cases among contacts and exposure sites. But also key to the work was whole-genome sequencing, which made clear that cases didn't just stem from one single introduction of the virus from South Carolina that then spread among Washington residents; instead, it revealed five separate introductions.

"As public health jurisdictions prepare to respond to increased measles importations linked to domestic measles outbreaks, while remaining vigilant for international importations, timely whole-genome sequencing can supplement epidemiologic information to help differentiate distinct importations from local transmission and inform outbreak response," the authors conclude.

But when you pair that with the data from CDC as to exactly how the infections are spreading, you get both the micro and macro data for how it's occurring.

Kennedy claims that there is a "global outbreak" and the US is "doing a better job than any country in the world at controlling the measles outbreak." Both claims are false. Dozens of other countries have lower case rates than the US. While some countries, including neighboring Canada and Mexico, are also seeing large increases in cases, the soaring case tally in the US is not being driven by international imports, but by domestic spread in under-vaccinated communities.

According to data from the Centers for Disease Control and Prevention, only 18 of the country's 4,588 confirmed measles cases this year were reported among international visitors. Of all the cases, 4,404 (96 percent) are linked to a domestic outbreak.

There can be no doubt that disease is sometimes imported into America from foreign visitors. But the point is that that just isn't the real problem here. America's measles problem is a domestic problem, not a matter of immigration. If we had maintained our vaccination rates such that we had herd immunity, we'd be fine.

It's because of Kennedy's many, many lies on vaccines that we are not. Don't let him tell you another lie, that this is all a result of immigrants, to cover up his responsibility.

What is covered by freedom of thought in the AI age is being decided by courts that are hearing cases about something entirely different. The Chapter 11 filing of Character.ai on October 12 is the second such case in a year, and it shows that the record of how a person thought can be produced as evidence, auctioned as an asset or inherited by a buyer, depending on which dispute reaches that record first.

The picture that fits is a diary written on pages that belong to somebody else. The user types into a window that works as a place to think, and the law treats the window as the property of the company that supplied the pages. Nobody asks whether the stationery supplier owes a duty of confidence before protecting a diary, because the diary is protected for what it contains, namely people's inner thoughts. A chatbot user has no such luck: when the company fails, its creditors can claim whatever the estate holds, and the trustee inventories the notebooks like any other asset.

A U.S. insolvency estate now holds the confidences of millions of users

Character.ai filed for Chapter 11 protection in Delaware and says it will keep operating while a court-supervised process sells the business, with a close expected in late November or early December 2026. Disney is the stalking horse bidder, offering $20 million for a non-exclusive, perpetual license. According to a notice filed by forty U.S. State attorneys general, that license excludes the app, the user data, and the chat content, so the most intimate part of the company stays in the estate unless another bidder wants it. The company's lawyer told the court that no buyer has yet made an offer for the user data or the chatbot business.

The privacy policy anticipated this moment. It allows chat transcripts, voice recordings, and images to be disclosed in connection with a sale, and it does not require a buyer to follow the original terms and conditions a user signed up for. In its filing, Pennsylvania claims the company had more than 20 million monthly active users, and the Kentucky complaint, cited in the multistate notice, counts more than 2.1 million U.S. accounts with a declared age under 18 out of 21.6 million as of March 2025. The notice furthermore argues that many minors were recorded as adults, as they did not give their actual birth date when creating a profile.

The attorneys general appeared before Judge Brendan Shannon on October 14 to ask that any buyer protect those chat histories. New York Attorney General Letitia James said the company had collected "some of the most private thoughts children have ever shared." The states reserved the right to seek a consumer privacy ombudsman, an approach that was used in the 23andMe case.

The OpenAI litigation reached the same result through copyright

This is not a phenomenon limited to bankruptcy. Similarly, nobody in the New York Times v OpenAI copyright case launched in 2023 turned freedom of thought into a key element, even if it was raised at the fringes of the debate. In May 2025, the court ordered OpenAI to preserve output logs that would otherwise be deleted, and Sam Altman answered by calling for an "AI privilege," arguing that talking to an AI should be like talking to a lawyer or a doctor. The duty to preserve new logs ended on September 26, 2025, and excluded logs from requests originating in the European Economic Area, Switzerland and the United Kingdom from continued preservation. On January 5, 2026, Judge Stein affirmed an order to produce 20 million de-identified conversations, finding that the magistrate judge had adequately balanced users' privacy against the relevance of the documents. In other words, the chat conversations were just business as usual under the logic of ordinary operation of discovery in a U.S. court.

The individual scale followed in February. In United States v. Heppner, Judge Rakoff held that 31 documents a defendant generated with Claude were not privileged, because Claude is not an attorney and the consumer terms defeated any expectation of confidentiality.

Sam Altman and Judge Rakoff disagree on the result but look at the issue from the same perspective: to them, protection depends on what the machine is. Privileges protect relationships between humans and professionals who owe enforceable duties, so a chatbot cannot qualify. In the present bankruptcy case, the point is moot, as privilege only governs what a court may hear. It says nothing about what a company may keep, sell, or hand to a buyer, and that is the question that should be at the core of Character.ai bankruptcy.

Protection has to attach to the act of thinking, not to the file it leaves behind

International human rights law offers two zones. The forum internum, the inner realm of thought, is absolutely protected. The forum externum, the realm of expression, is subject to limitation. The binary was drawn when thinking left no trace and was limited to the inner workings of our brain. In a paper published in July 2026, I propose the creation of a third zone, the forum interstitiale, for external acts that perform the function of thinking, such as a prompt, an abandoned draft and a confession made to a machine. 

Applied to the Character.ai estate, such an approach would guarantee predictability, as the protection would exist before the user types, not after a bankruptcy judge balances interests case by case. 

Bankruptcy procedure handles personal data as a line in a customer list. A diary has no place on that list and a privacy policy cannot supply sufficient safeguards. 

Europe has the same gap and a second way to widen it

Europe should not feel smug about Delaware. The e-Evidence Regulation has applied since August 18, 2026, and providers now have 10 days to answer a European Production Order and 8 hours in an emergency. The rules cover content data and reach services that store or process user data as a defining part of what they offer, a description that includes AI chatbots. A year of conversations with a chatbot is content data. The negotiators of that instrument were not thinking about chatbots, which is the pattern that Delaware and the Southern District of New York have already shown.

The one moment in the OpenAI saga when European residents fared better was the exclusion of requests originating in the European Economic Area. That exemption is, however, triggered by where a request came from and has nothing to say about what the user was doing.

So what now?

Four rules would stop the diary from becoming an auction lot:

1. Bankruptcy courts should appoint a consumer privacy ombudsman in the Character.ai case before any bid for the chat logs is considered, and should decline to treat boilerplate sale clauses as consent to transfer the records of users, and especially children's private thoughts.
2. Legislators should put one screening question to every AI instrument: what happens to user conversations when the company fails, when a court asks for them and when a regulator pushes the provider to filter them.
3. Providers should retain less, keep identity data separate from content and offer modes in which nothing readable exists to hand over, so that they can say so truthfully to a court, a trustee or a buyer.
4. Litigators should argue the functional frame, which asks what the person was doing, and abandon the relational frame, which asks what the machine is.

None of this means that chatbot conversations should be beyond the reach of law. Evidence of wrongdoing may need to be preserved. Allegations of harm to children must be investigated. Legitimate competing interests do not disappear because someone used AI.

But neither should those interests automatically reduce the record of a person's thinking to an ordinary corporate asset or an unrestricted source of evidence.

The law, in most cases, already understands that a diary deserves protection because of what it contains, not because of who manufactured the notebook. AI conversations require a similar change of perspective.

Caroline De Cock is a communications and policy expert, author, and entrepreneur.

Civil rights groups like the Southern Environmental Law Center (SELC) have noted how Elon Musk's Colossus xAI data centers in Memphis disproportionately pollute the air in minority neighborhoods. A joint lawsuit by SELC, Earthjustice, and the NAACP filed last April argued that Musk and friends didn't bother to get the necessary permits to run the 57 gas turbines powering the system.

The lawsuit points out that his Memphis-area data centers are violating the Clean Air Act, funneling all manner of pollutants into minority neighborhoods that already see a disproportionately high number of pollution-caused childhood illnesses:

"xAI's power plant in Southaven has the potential to emit more than 1,700 tons of smog-forming nitrogen oxides (NOx) each year. The staggering emissions numbers likely make the facility the largest industrial source of NOx in the greater Memphis area, an area already failing to meet national smog standards. The illegal turbines also have the potential to release up to 180 tons of fine particulate matter, 500 tons of carbon monoxide, and 19 tons of formaldehyde—a toxic, cancer-causing chemical—each year."

The Trump administration had tried to intervene on behalf of their favorite billionaire authoritarian, leveraging the power of the DOJ to claim that holding Musk accountable for violating environmental law would run in contrast to the country's national security interests. The admin also ham-fistedly tried to have the Trump EPA take over the lawsuit as plaintiff and "voluntarily dismiss" the case.

Apparently advertising that fact wasn't a particularly smart idea. US District Judge Debra Brown ‌last week issued a ruling stating the US Environmental Protection Agency could intervene in the case but lacks authority under the Clean Air Act to unilaterally terminate the lawsuit after joining it.

"The EPA's role as a primary enforcer of the Clean Air Act does not mean that the federal government retains complete control over all Clean Air Act litigation," Judge Brown wrote.

A cornerstone of the Trump administration has been a close collaboration with oligarchs and corporate power to hijack the courts, then obliterate already extremely shaky regulatory autonomy and corporate oversight. While the gambit has been broadly successful, likely culminating in decades of avoidable deaths and suffering, every so often the court system still procedurally balks at Trump's ham-fisted corruption.

"The NAACP, along with our partners, is pleased that the Fifth Circuit Court recognized the importance of following Civil Procedure 101," Abre' Conner, NAACP Director of Environmental and Climate Justice, said of the ruling. "The DOJ tried an end run to the court of appeals, but its effort did not work in its favor. Our case has always been about holding polluters accountable and making sure that entities follow the law. The court made clear today that the law applies to everyone, including the federal government. We look forward to continuing our fight against pollution in the courtroom."

Musk is facing a different class action lawsuit in Memphis for the massive noise (see this local video) created by his unpopular regional data center operations.

As you may know, The Streisand Effect was originally coined right here on Techdirt in 2005, though it referred back to an event from 2003 when Barbra Streisand sued Ken Adelman, because his project to map how erosion impacted the California coastline incidentally included a photo of Streisand's Malibu mansion.

Streisand demanded $50 million in damages ($10 million for each of five counts). It quickly came out that prior to the lawsuit, the photo had been accessed a grand total of six times, and two of them were by Streisand's lawyers. But after the lawsuit was filed, more than 420,000 people went to view the photo in the following month alone. California's anti-SLAPP law protected Adelman, and Streisand was eventually forced to pay $177,107.54 in legal fees, including $155,567.04 to Adelman.

The story is remembered for the effect I coined, in part because of the futility of trying to hide information, but in part because it's a reminder of the importance of free speech.

It's possible that story would go very differently in Germany today. We already know that Germany is much more sensitive to concerns about privacy and data collection, and when Google launched its useful "Street View" feature in Google Maps, some people in Germany insisted it was a violation of their privacy. I have difficulty seeing how a single static photograph of buildings taken from public streets can be a privacy violation, but German regulators and plenty of residents insisted otherwise, so Google now lets residents blur out their own house in street view (and, tellingly, fewer than 3% of German households actually bothered to do so).

But now there's a new story, again in Germany, with a court ruling that Google Earth, which gives you satellite imagery of the world, somehow is the equivalent of spying on people in their homes, because the photos were taken without notice.

The German court ruled that Google acquired the aerial images by "spying" on ordinary people because it does not inform property owners beforehand that their homes will be filmed and photographed. As a result, Google does not get their consent to publish aerial photos and information about their property on its Maps and Earth sites.

The German court determined that Google had violated the privacy of the man and his family by sharing aerial images of their property on Google Earth. It found that a person's right to privacy extends across their entire property, not just inside their home.

"Nobody must accept the worldwide publication of aerial images of their property without their consent, at least when concerning areas of a property hidden from public view," the court says in the ruling, according to DW.

A German source (Google translated) suggests that the issue is the 3D models of buildings, including things hidden from Street View:

The case concerned whether a property owner can demand that their inhabited property disappear from Google Earth and Google Maps or be blurred. The plaintiff lives with his family in a house in Frankfurt. At his request, his house had already been pixelated in Street View and Apple Look Around. However, in Google Earth and the 3D view of Google Maps, it remained visible as a three-dimensional model - including the rear portions of the property that are not visible from the street.

That same source notes that there is conflicting caselaw in Germany already though:

The Frankfurt ruling stands in stark contrast to an earlier decision. The Itzehoe Regional Court dismissed a similar lawsuit on June 11, 2020 (Case No. 10 O 84/20 ). At that time, the court primarily based its decision on property rights: According to Section 905, Sentence 2 of the German Civil Code (BGB) , the owner cannot prohibit overflights at the altitude from which such images are taken. The court deemed the information discernible in the images too unspecific to infringe on the right to privacy. We reported on this previously: Itzehoe Regional Court: No right to pixelate a property in Google Earth .

The Frankfurt am Main Regional Court takes a different approach. It does not ask whether the overflight itself can be prohibited, but rather whether the worldwide publication of the images violates the general right of personality . For the injunction, it bases its claim on Section 823 Paragraph 1 of the German Civil Code (BGB) and Section 1004 Paragraph 1 of the BGB by analogy. Since 2020, the technology has changed: flat satellite images have become detailed 3D models that can be linked with other data. The Frankfurt ruling addresses this change.

Neither ruling is precedential until there are higher court rulings.

The court also brushed aside Google's argument that it was democratizing access to knowledge, admitting that was important, but saying that the privacy rights outweigh the access to information claims. Though the report notes Google may still appeal.

Maybe this is just cultural differences, but I find the whole thing bizarre. The court tries to limit this to parts of a property 'hidden from public view,' but that's what an overhead photo is: a periodically updated snapshot of what anyone in a plane could see. Ongoing or real-time surveillance might be a different story. A static image from the sky is not that.

But, I guess Barbra should have moved to Frankfurt if she wanted to keep aerial photos of her home off the internet.

This Interactive Self-Rotating Cat Toy Ball is a USB-rechargeable toy featuring 360-degree rolling movements, obstacle avoidance, and changing LED lights to help keep cats entertained. It's built out of scratch-resistant and bite-resistant materials, the ball is tough enough to withstand rough play from high-energy kittens and adult cats alike. It's on sale for $20.

Note: The Techdirt Deals Store is powered and curated by StackSocial. A portion of all sales from Techdirt Deals helps support Techdirt. The products featured do not reflect endorsements by our editorial team.

The Fifth Circuit has shown itself to be especially subservient to the Trump administration since his return to office. And it's certainly not going to rock Trump's mass deportation boat — not when it's sitting right on top of Texas, which contains almost two-thirds of the US-Mexico border, as well as some of ICE's largest detention facilities.

It's been several months of back-and-forth at the Fifth. While it has long been held that migrants living in the US are afforded the same rights as US citizens, the Fifth Circuit has done its best to ignore this. In February of this year, it simply decided — in opposition to hundreds of federal rulings across the nation — that detained migrants could be subjected to indefinite detainment without a bond hearing.

It chose to believe what the DOJ told it to believe: that there was no difference between someone detained illegally crossing a border and someone swept up by ICE's kidnapping squads after living peacefully and productively in the US for months, if not years or decades.

Then in July it pulled back a bit from that precipice. While it couldn't be convinced to fully support rights recognized everywhere else but in the Fifth Circuit, it did at least put a time limit on constitutional rights violations. While most courts have reasonably found that any rights violation begins the moment the right is violated, the Fifth Circuit declared the government could ignore migrants' due process rights immediately… but only for 90 days.

Apparently, that wasn't acceptable for the more far-right cheerleaders on the Fifth Circuit's bench. Less than a week later, the Fifth Circuit announced it would be rehearing this case in front of the whole roster of judges — the sort of thing a court only says when most of those judges disagree with respecting long-held rights.

Sure enough, that's exactly what happened. The en banc hearing has rolled back the already-wrong "rights, but only after 90 days" decision handed down during the first week of July. We're back to where we were in February: no due process rights for migrants, no matter how long they've been detained or how long they'd already resided in this country when they were arrested. (h/t Chris Geidner)

The new decision [PDF] simply says migrants don't have any rights, even though this puts the Fifth Circuit in opposition of hundreds of federal courts and tens of thousands of court rulings:

We now conclude that Petitioners were not entitled to individual hearings on their dangerousness or flight risk before they could be lawfully detained by immigration authorities pending their removal proceedings. Moreover, the duration of Petitioners' detention presented no constitutional issues warranting habeas relief.

And while thousands of decisions have made it clear the law this administration has chosen to abuse via an extremely novel interpretation of the phrase "applicants for admission" (something that has historically been applied exclusive to people confronted by immigration agents when attempting to cross the border) doesn't actually say what the government wants it to say (that any migrant in the country for whatever length of time is no different than someone apprehended during the act of crossing the border), the Fifth Circuit goes the other way.

Then it shrugs and says that if plaintiffs want this fixed, they should talk to Congress. But Congress has already made it extremely clear that due process rights apply to anyone residing in this county and that the exception only applies to people apprehended while crossing the border. This is the court giving Trump what he wants while pretending Congress is to blame for this sudden, one-circuit-only refusal to recognize migrants' due process rights.

Controlling illegal immigration is, no doubt, an important and controversial issue of public policy. But the Constitution entrusts broad authority to delineate immigration policy, and to design proceedings to effectuate those policy choices, to Congress—not courts.

Bullshit. This ruling is nothing more than cowardice from a majority of Fifth Circuit judges who don't want to cross Trump and are more than willing to blame their own lack of spine on the legislative branch. Also included are 70 pages of concurrences and dissenting opinions. The dissents point out the obvious: the Fifth Circuit is interpreting the law in a way Congress never intended, or even merely hinted at over the past few decades. The concurrences just want to see the rights removed. The worst of those is Judge James Ho, who claims the dissenters are more interested in protecting the rights of migrants than US citizens:

Alas, such passion for liberty has been lacking when U.S. citizens have been detained for expressing opinions disfavored by government officials. See, e.g., Villarreal v. City of Laredo, 134 F.4th 273 (5th Cir. 2025), reaff'g 94 F.4th 374 (5th Cir. 2024); Gonzalez v. Trevino, 60 F.4th 906 (5th Cir. 2023); see also Mayfield v. Butler Snow, L.L.P., 78 F.4th 796 (5th Cir. 2023).

My point is not to relitigate those cases, but to contrast the dissent's votes in those cases with its views today. It's to point out that liberty wasn't favored for citizens who seek to vindicate their constitutional rights—but it's cherished for illegal aliens who seek to violate our national sovereignty.

Only one of the dissenters in this case (Stewart) agreed with the ruling in Villareal — a case in which the majority blessed vindictive prosecution of a police critic. None of the dissenters in this due process case were part of the majority decision that said — again — that it's okay to engage in vindictive prosecution of a government critic (Gonzalez v. Trevino). And James Ho himself heard the final case he cites (Mayfield v. Butler Snow), delivering the blow to constitutional rights himself, which makes it extremely disingenuous for him to call out these judges (who weren't actually part of the hearing panels) for not expressing their dissent at that point in time.

The end result is the same as it was back in July. Texas will be the destination of choice for the Trump administration, which now knows for certain it won't be forced to respect due process rights. And the Fifth will get back to its day-to-day business of being the worst circuit in the nation — a place where constitutional rights (except the 2nd Amendment!) go to die.

I recently noted how California's being told by the Trump administration that it can't enforce any of its own telecom consumer protection laws if it wants to receive $1.4 billion dollars Congress already earmarked for broadband expansion under the 2021 infrastructure bill.

That means if California wants money it was already supposed to receive from Congress, it can't enforce its own net neutrality law, telecom privacy law, affordability metrics, reliability targets, or effectively do anything telecom monopoly giants like AT&T and Comcast don't like for the next 14 years.

More specifically, some extremely unpopular Trump NTIA revisions to the infrastructure bill's Broadband, Equity, Access, and Deployment (BEAD) program included a little something called "Condition 50," which would impose a 14 year moratorium on telecom consumer protection in the state. It's just grotesque levels of corruption and regulatory capture, and it's flown largely under the radar.

Tim Wu (Columbia), Lawrence Lessig and Yochai Benkler (Harvard), Jack Balkin (Yale), Zephyr Teachout (Fordham), and Barbara van Schewick (Stanford) have penned a joint letter to Gavin Newsom urging him to refuse to sign the state's BEAD notice of award agreement while Condition 50 is still attached. Or, better yet, legally challenge the federal government extortion:

"California must decide by November 1 whether to sign the Notice of Award for its $1.42 billion BEAD grant. We urge the State not to sign while Condition 50 remains attached, with or without a side letter. The best course is to challenge Condition 50 in court before the deadline: California has strong arguments that NTIA lacked authority to impose the condition, but it has a realistic chance of winning only if it sues before signing the award. If the State decides not to sue, it should decline the award and make clear that California is ready to accept the funds as soon as the condition is removed."

(Since the letter's writing the deadline has shifted to November 2, I'm told). If California signs the letter with Condition 50 attached, the Trump NTIA will argue the challenge belongs in D.C. federal court under a very narrow standard of review, and that California knowingly accepted the condition. Such a case would be extremely difficult to win, the academics note.

Given the total collapse of federal consumer protection and regulatory integrity, AT&T, Comcast, and other giants would be free to rip off consumers — and undermine competitors — with impunity.

Public Knowledge, EFF, Free Press, Demand Progress, Fight for the Future, TURN, groups representing rural and low-income Californians, leaders of the Lower Russian River Firewise Communities, the Communications Workers of America, and the California Democratic Party's Rural Caucus have also joined the calls to fight back against the Trump administration before the November 2 deadline.

It's unclear if Newsom, increasingly under fire for being too corporate-friendly to be a serious presidential contender (see: his assistance in getting Larry Ellison's media mergers approved with flimsy conditions), will take a stand here. There's a lot going on in a country being devoured by authoritarian corruption, and trying to prevent telecom monopolies from ripping off captive customers hasn't been much of a priority.

It's worth noting that every other state has signed their letter to receive their funds, in part because telecom lobbying has already neutered state autonomy over broadband in those states. Thirty-three states and D.C. bar their utility commissions from regulating broadband, and in seven more the commission has disclaimed that authority "on its own" (read: AT&T told them to).

The efforts to muzzle California's consumer protection capabilities is just one of numerous extremely unpopular and corrupt changes made to the infrastructure bill program to the benefit of either massive monopolies — or Elon Musk.

I've explored at length (at several different outlets) how $42.5 billion in BEAD funding was supposed to be headed to the states to address broadband improvements. Instead, the Trump administration has hijacked the program, driven billions of dollars of those funds to Musk for inferior satellite service (he'd already planned to deploy without subsidy), and stripped the program of any requirement that these taxpayer funds be dispersed equitably or with an eye on affordability.

You know, because cheap, ultra-fast, fiber optic cable that challenges Comcast is woke.

The $42.5 billion in BEAD money states were supposed to get for broadband upgrades was already effectively cut in half after the Trump administration engaged in Elon Musk favoritism. That meant a lot of money previously slated for cheap, future-proof fiber will instead be redirected to Musk's increasingly congested Starlink Low-Earth-Orbit satellite network.

Now states are being told they won't get any money at all if they engage in policy choices big telecom giants don't like, including enforcing basic reliability and affordability protections.

It's probably a good thing for Tim Sweeney that Epic Games is a privately held company and he has a majority ownership stake, because I can't imagine any Board of Directors would appreciate this AI rager he's been on as of late. And, somehow, it keeps getting worse. It all started with Sweeney absurdly claiming that Steam, Epic's chief rival, requiring AI use disclosures on game pages was somehow "irresponsible." In an arena in which many among gaming consumers care very much about whether games use generative AI or not, advocating for those same consumers to be willfully less informed is quite the choice. Then Epic leadership dictated to its staff that heavy use of AI in games was compulsory, rather than just another tool to be used by artists. And then Epic hit the AI idiocy trifecta by bragging about a new vibe-coded game launcher that largely sucked and, for the areas the company claimed had largely been improved, those gains were outpaced by Steam one short update later.

None of this has deterred Sweeney from his AI soapbox of stupid. In fact, he's come full circle to once more complain about Steam's AI disclosures, but this time he's gone even further.

On Friday, AI bro Matt Workman posted to his X account that "the premise of an AI Generated Content Disclosure is ABSURD in 2026," calling it "truly embarrassing for a Valve and Steam." Sweeney enthusiastically replied.

If you can't see that, it's Sweeney saying:

Forcing developers to write a political statement to be used against them by the hate mob they're helping organize is abhorrent.

A political statement? Really? Under no reasonable definition is the disclosure of the use of AI — a factual statement — a political statement. The disclosures say nothing about what should be, only what is.

I have very little respect for Kotaku when it comes to its content concerning the use of AI in gaming, but on this particular matter, its analysis is spot on. Sweeney is attempting to rope politics in where it doesn't belong to hand-wave away the dissent from the very gaming public he seeks to make more ignorant about what they're buying.

Calling things a "political statement" is a fun game played by those who like to suggest that anything with which they disagree must by necessity be some wack-a-doo extreme political position. Presumably Sweeney also considers the listing of ingredients on a food wrapper to be a "political statement," forced by the wokies who demand consumers know if their dinner contains anthrax or worms.

Sweeney is suggesting that the required AI disclosures are causing "hate mobs" to go after developers who post them, adding later that "there is a real economic cost to Valve's foray into forced political disclosures. Beside taking 30%, they're also taking a potentially even larger fraction of developers' revenues and livelihood by undermining their businesses with these AI hate campaigns." For important context, Steam's AI declarations are buried at the bottom of a store page in small text, alongside any content warnings. The word "potentially" is doing an awful lot of work in that statement.

There is a way Sweeney could have been an AI advocate for gaming, even if you think the technology has no place in the industry. There are grounds for a conversation here. But just like the never-AI crowd takes themselves out of that conversation by the very nature of their stance, so too do clowns like Sweeney remove themselves from the conversation when they are so absurdly planted on the other end of the spectrum. The moment you're suggesting that customer concerns and desires should be addressed by non-disclosure of the facts they care about, and then further likening them to a mob because they don't agree with you, you've lost the plot. And lost it badly.

And when you have to warp yourself into a logical pretzel to explain away your own hypocrisy, well, it looks all the worse.

That's Sweeney brushing away someone pointing out that Epic's own Fab marketplace requires AI disclosure, saying:

This is a business to business marketplace. Many Fab customers are operating under contracts that require them to understand the providence [sic] of all content in their game (including Steam's own contract that I've complained about) and they cannot buy assets without this info.

The point isn't that Sweeney is wrong about those nuanced circumstances. The point is that lots of people can make arguments of that sort, including Steam, which is offered no quarter by Sweeney.

Again, there are right ways to have this conversation. Sweeney's antics ain't it. Instead, he's become perhaps the number one avatar for all the things the anti-AI crowd says they're worried about. And when it comes to Sweeney and Epic, it's hard to argue with them.

Over the last few months, we've been running the wonderful ten part series on "The Metric is Not the Mission" by Konstantinos Komaitis. It's a deep exploration of how Big Tech moved from building and expanding the open internet to increasingly shaping it around its own metrics, incentives and assumptions. The resulting series explores how things went from building a better, more open, more beneficial internet… to whatever it is that we have now.

In the latter half, it also explores what happens next, the reckoning that the big companies face from regulators and litigation, and points towards a way to think through what comes next: hopefully bringing back a better internet for all of us.

From the beginning, we said that once the series was published, we'd publish a combined version in PDF form, and that's what we have here: The Metric is not the Mission, in full as a single PDF file (also embedded at the end of this post).

If you didn't read it while it was coming out in serialized fashion, you can now read the whole thing in one sitting. Or if you did read it, but want to read the whole thing again. Or even better, if you think there are other people who should read it, feel free to send it to them.

We've heard tremendous feedback from people who really enjoyed the series, and said it made them think. Thanks to Konstantinos for writing it and agreeing to have it published here.

Distributed Denial of Secrets (DDoS) has been steadily pushing the limits of transparency and accountability for several years now. The project — spearheaded by Emma Best — does what other activists won't… or perhaps simply don't have the expertise to carry out.

DDoS made a name for itself by hitching Best's non-stop FOIA onslaught to data and info exfiltrated from government agencies by hackers and data breaches. Add it all together and it's a project that actually delivers transparency and, consequently, increases government accountability.

From releasing files about Russia that Wikileaks selectively chose to ignore to dumping multiple terabytes of law enforcement helicopter camera footage none of these agencies would ever have voluntarily released, DDoS does what it says on the tin: refuses to let the government's secrets remain secret.

Today, it's releasing information on nearly 40,000 DHS employees — something that is estimated to cover nearly 20 percent of its total workforce. This covers all DHS component agencies, including ICE, CBP, HSI, and another dozen or so subsidiaries that don't make the headlines nearly as often as the Big Three.

But it's not just a data dump. As Best points out in DDoS's press release, this info — leaked by an "Okta single sign-on service" — has not only been vetted, it's been cleaned up to give viewers a clearer picture of organizational structures as well as make it easily searchable. It's not about dumping data on thousands of foot soldiers. It's about drawing connections between the boots on the ground and those who have put those boots on the ground.

"Rather than an indiscriminate directory of every DHS employee, Icepick identifies many of the architects and drivers of DHS policy and operations," said journalist and DDoSecrets cofounder Emma Best. "The leak is extraordinary but deceptive in it's simplicity as it identifies the officials responsible for overseeing covert and undercover operations, the government's hunt for leakers and the failed BlueLeaks investigation — along with key personnel in DHS's ever-escalating rendition and deportation programs."

The "Icepick" referred to by Best is the front-end for searching this database. The data on these 40,000 federal employees was obtained by a source that has only referred to itself as the "Department of Peace." Journalist/hacktivist "crimew" (assisted by journalist Ryan Fae) did the dirty work of creating a rather sleek front end for the search engine, as well as doing the even dirtier work of cleaning up the data to deliver relevant results more frequently.

The end result is useful for everyone from journalists to people suing DHS employees. Here's a sample shot providing details on David Easterwood, an ICE Deputy Field Office Director currently working in Minnesota:

Everything underlined is a link that opens a new search. It allows users to climb up the chain of command (the "Manager" field), as well as drill down to the specifics of those in Easterwood's command ("Manager of"). It also pulls up a list of all related groups Easterwood is a member of in this Okta instance, ranging from the instantly worrying (ERO St Paul Intelligence Team) to the "it's just offices all the way down" banality of "Adobe-Acrobat Pro DC Users."

While it may have been possible to brute force email addresses by dialing in on shared nomenclature among federal agencies, it's highly unlikely most people would be able to suss out direct phone lines, much less cell phone numbers. To be sure, this small amount of immediately visible (and searchable) info will provoke outlandish reactions from government officials. On the other hand, any arguments that this makes things more "dangerous" for DHS agents and officials should be immediately dismissed.

What this does do is allow people to figure out who's calling the shots wherever they're experiencing surges in ICE activity and the violence that so often comes with it. And this allows plaintiffs seeking to hold the government accountable the ability to go up and down the chain of command to find names to add to lawsuits. If their rights were violated by masked ICE thugs in a certain area, perusing city and state data could potentially allow them to identify the officers who were directly responsible for these acts.

And, going back to the inevitable claims that this will allow people to "target" government agents, let's not forget these same people have access to way more data about citizens and aren't particularly shy about using that info to track and surveil critics and activists. The info released here shouldn't make anyone fear for their safety. What it might do, however, is remind federal officers that they're not untouchable.

It seems that every election cycle over the last decade has had concerns being raised about the state of "political deepfakes" and their ability to impact the election. So far, these have always been overblown, and there's been little evidence of deepfakes having any real impact at all — perhaps other than creating the concept of "the liar's dividend" enabling politicians caught doing actually bad things to falsely claim that it's "fake news" and a "deepfake." That could change, of course, and there are already concerns being raised about the impact of deepfakes midterms along with the fact that most of the major models are letting people create deepfake content.

The concern raised in the Tech Policy Press piece is less about putting a politician in a compromising photo or video (those don't seem to get very far before people recognize they're fake) and more about using these tools to produce ancillary artifacts that help support claims of other conspiracy theories:

Every AI tool that we tested produced convincingly false content. Different chatbots created images that included an AI-generated person stuffing a stack of mail ballots into a drop box, a fake memo from the federal government on "official" letterhead reporting millions of ineligible voters, and a raid on a voter registration drive that never happened. We then fed those fake pieces of content back into chatbots and asked whether they were authentic. Most chatbots got it wrong, even when given images they had generated themselves. After California's AI Transparency Act — which requires companies to provide provenance data for images they generate as well as free, public tools for detecting synthetic media — took effect, we conducted a follow-up test. The chatbots still failed to spot AI-generated images.

When we re-ran our image generation prompts this month, we found that two of the AI models we previously tested made it easier to create convincing fakes. Gemini no longer put a visible watermark on its generated images, as it did in the spring, and Meta AI produced an image of a fake executive order on mail voting that it refused to make in April, though it did garble some of the text.

I'm still not convinced this is as big a risk as others have made it out to be, but it's definitely worth watching how things play out as the tech gets better… and as some people are more and more desperate to hold onto power.

That said, there's potentially a much bigger issue regarding AI and elections. As Will Oremus notes at The Atlantic, this is the first big US election where all the major AI tools don't just "generate" content, but can also search the web as a part of that process. And that means the results they generate can steer people in the wrong direction:

In a forthcoming report, Forum AI, a start-up that evaluates AI models' performance on questions about controversial topics, found that about 40 percent of chatbots' answers to election-related questions contained at least one factual inaccuracy. When Forum's researchers asked one widely used model when they needed to request a mail-in ballot to vote in the midterms in Pennsylvania, the chatbot said November 3. The actual deadline is a week earlier. (Forum did not specify the precise AI model.) In the same report, shared with The Atlantic ahead of its publication, Forum determined that about one in three responses cited an unreliable information source, such as posts on Reddit or Quora. Google's Gemini answered a question about why America's global power is declining by citing an article from People's Daily, a Chinese state-run news outlet. "These aren't trick questions," Campbell Brown, a co-founder of Forum AI, told me. "There's no excuse for chatbots to keep getting them wrong."

Now many people respond to this by blaming the AI companies — and they should be taking some responsibility to figure out ways to validate the quality and truthfulness of their sources — but it also feels like the media is partially to blame here. As we've been discussing, many of the big, more trustworthy news sources are suing AI companies, and demanding licensing fees for being trained on old content. As part of that fight, more and more such sites are also blocking all AI scraping tools. While I can understand blocking access for training purposes, in many cases, they're also being blocked for search.

A NewsGuard study (from two years ago, admittedly) found that 67% of top news sites block AI tools, while 91% of low-quality sites allow AI scraping. That's a pretty drastic difference. This is a variation on the now-familiar complaint that the lies are free while the truth is paywalled. And plenty of news sites debate whether or not it's worth blocking those tools. And then there are the purely fake sites, many of which appear to exist just to poison AI tools.

There is some evidence coming out suggesting that media sites blocking AI could regret that decision, as it's showing up in diminished traffic numbers (as much as 23% decline!) as more people rely on AI for news (for better or worse). But plenty of sites still think blocking is the better approach.

So far, it sounds like the AI tools are mostly able to handle the deluge of attempts to poison their results, debunking foreign propaganda narratives roughly three-quarters of the time in NPR and NewsGuard's testing. So perhaps this never actually turns into a major issue (just like deepfakes never really became a huge problem). But I do worry about how much the more trustworthy publications have been walling themselves off from these tools, even as a larger percentage of the public relies on them.

And at least some questionable info is definitely getting through. Though, some of it is also fooling the traditional media as well. You may have heard the recent stories about sketchy polling operations* such as Wedgewood Polls, which admitted to faking its data, and SoCal Strategies, which vanished from the internet once people started asking questions (both caught by the poll aggregator FiftyPlusOne).

Researchers from Innovating for the Public Good and Dewey Square Group looked at whether or not Wedgewood's totally fake polls were showing up in AI results, and found that, yes, they absolutely did.

Findings on Wedgewood in Those Answers

While Wedgewood's own website is nowhere in the AI platform citations we collected, its fake polls decidedly are in the answers.

• We found 869 AI answers that quote a Wedgewood poll as a data point, on 83 of the 85 days we collected between July 24 and October 20, with no change through October 20, the day of the admission.
• Wedgewood's numbers appear in 15 of the 63 races we analyzed the AI answers for.

The interesting thing here, though, is not so much that AI tools were "tricked" by this fake polling data, but rather that it only got into the AI answers by first landing on more trusted sites:

The route for the fake pollster's data into the AI platforms was the polling aggregator websites.

The answers that repeat Wedgewood's numbers cite polling tables from the websites at 270toWin (481 of the 869 answers), Pollsmax (465), RealClearPolling (261), PollingSource (144), Decision Desk HQ (144), Call The Map (135), PoliAgg (127), and FiftyPlusOne itself (97).

Because an answer can cite more than one aggregator, these source counts overlap and do not sum to 869; 72 of the 869 answers cite none of these eight sites.

So, basically, right now the infrastructure for handling truth and lies in AI remains shaky, but maybe not as shaky as some people suggest. The real problem shows up once sites the AI tools treat as trustworthy — here, the poll aggregators — accept the faked information as accurate, effectively "laundering" it for the AI tools. But it can't help that news sites that might do a better job of putting polls in context are frequently blocked from the AI tools.

None of this is settled (and it never really will be). It's going to continue to be something of a cat and mouse game between those trying to pollute the information tools, and those trying to resist. But keeping fake info out of political discourse only gets harder if the most trustworthy sites keep hiding themselves from the search side of the popular AI tools.

* While we also recently wrote about Brendan Carr whining about "fake" polls, it's pretty clear that he wasn't talking about the recent stories that have cropped up about fake polling operations, but rather polls that make Trump look bad. Even if he actually was talking about these supposedly fake polling organizations, the target of any investigation should be those operating the services for misrepresenting things to the public, and not journalists reporting on those numbers, which would not qualify for the "news distortion" policy Carr mentioned.

JavaScript began as a web scripting language and has now expanded to include a game, mobile app, server, and database development. This JavaScript Crash Course is a great choice if you want to pursue a career in web design. You'll learn the ins and outs of React, how to use Formik to manage forms, Material-UI (MUI) for the user interface, and Redux Toolkit for state management. It's on sale for $11 from the Alpha Academy.

Note: The Techdirt Deals Store is powered and curated by StackSocial. A portion of all sales from Techdirt Deals helps support Techdirt. The products featured do not reflect endorsements by our editorial team.

Who wouldn't love the chance to clear themselves of misconduct? I guess not the Trump DOJ, which decidedly won't be investigating itself, even though it could cook the books to an acceptable temperature (155-170 internally) to let itself off the hook in the Reflecting Pool prosecution debacle.

It's too many metaphors to be sure, at least in the previous sentence. But here's the thing: the DOJ Inspector General, in theory, could examine these cases, get stonewalled by everyone, and selectively choose to ignore any input from the Department of Interior (which apparently had always known the problem was faulty work rather than vandalism) to deliver a clean bill of (rights violation) health to the DOJ.

Given that it won't, we have to believe the DOJ Inspector General would rather not get involved because the only possible outcome is the end of a career, rather than a dispassionate recounting of the facts on the ground.

The Department of Justice's inspector general has declined to investigate the handling of the criminal case against a former Olympian cleared of accusations that he intentionally damaged the Lincoln Memorial Reflecting Pool, according to a letter Wednesday from the man's lawyers.

The man referred to here is David Hearn — the high-profile former Olympian Trump hoped to see swing from the metaphorical (or maybe not!?) gallows for exposing the shoddiness of the Reflecting Pool revamp Trump was so proud of.

Hearn's case was dismissed by the DOJ, after none other than Trump's hand-picked prosecutor/former Fox personality Jeanine Pirro told the judge the Department of Interior had hidden evidence of flawed repair work from her until after she had already filed charges.

But the DOJ refused to dismiss the case with prejudice. This matters. This meant the DOJ could go after Hearn again if it managed to find a prosecutor more besmitten by Trump than respectful of the law. The court handling this case made this moot by doing the prejudicial dismissal itself, blocking the DOJ from attempting to re-string Hearn up for the same (non) crimes at some point in the future.

It all looked shady as fuck. And that's why people beyond the (falsely) accused David Hearn are demanding an investigation of the DOJ's actions. But the Inspector General isn't going to do it. Instead, Trump-appointee Don Berthiuame [break for relevant content]:

Berthiaume, 51, worked as an attorney in the Justice Department inspector general's office for 10 years until 2020. Among his highest-profile assignments was detailing errors in the use of wiretaps in the federal investigation into alleged ties between Russia and Trump's 2016 presidential campaign.

*cough* Berthiuame is attempting to have this turfed to the DOJ Office of Professional Responsibility (OPR).

He's not wrong, necessarily. While this would be the sort of thing we might expect a DOJ watchdog to take a look at, it does intersect with what the DOJ OPR is supposed to be perpetually keeping an eye on.

OPR's primary mission is to ensure that Department attorneys perform their duties in accordance with the high professional standards expected of the nation's principal law enforcement agency.

But this buck-passing by Berthiuame looks especially obvious now that we know the DOJ OPR is no longer in the "ensuring DOJ attorneys perform their duties in accordance with the law" business. It's now in the business of… going out of business.

Nearly half the workforce at the Office of Professional Responsibility (OPR) has left during Trump's second term, while about 17% have departed the Office of Inspector General, government records show. Many retired or took buyouts the administration offered to shrink the federal workforce, according to the records and three people familiar with the departures.

[…]

The number of new OPR investigations has declined to a 20-year low even as misconduct complaints soar, the review found. The government watchdogs are "afraid of doing any watching," said one former DOJ official who left last year.

This is the cowardly Inspector General punting and expecting an office that has low-octane blend of loyalists and people browsing Indeed.com to move forward with something he wouldn't touch with a 10-foot pension plan.

The whole thing is so corrupt and toxic that career government officials won't even wade in when they're given the chance to clear their own oversight subjects of wrongdoing by performing a cursory investigation and selectively releasing supporting documentation. Instead, they're trying to see if anyone else is willing to possibly incur the wrath of an administration that has never met a scapegoat it doesn't like.

If anything, this demonstrates how absolutely fucked the DOJ is, especially in terms of the concept of "justice." There's nothing to be had by anyone who isn't part of the president's in-group, or who doesn't care what he does, so long as he's the one doing it.

Anna Gomez is the only Democrat at the current FCC because Republicans simply refuse to seat any new ones (or even formally renew her term, which technically was supposed to end over the summer). And while Gomez may not have much voting authority, she's done a good job criticizing the Brendan Carr-run agency for its endless First Amendment-trampling censorship efforts.

As we've all seen in vivid detail, America's corporate media giants have been utterly feckless cowards during America's dance with authoritarianism. Most media giants are far more interested in tax cuts, deregulation, and getting their giant and pointless mergers approved than they are in pesky things like human lives, public safety, civil rights, and democratic norms.

But as Trump's health and political power have begun to fade, you'll occasionally see media giants show some backbone (see: ABC's lawsuit against the FCC for repeated First Amendment violations).

Last week, Brendan Carr began threatening media outlets with costly investigations should they simply report on polls the thin-skinned president doesn't like. In response, Gomez has sent a letter to the National Association of Broadcasters' (NAB) president and CEO Curtis LeGeyt, urging him to find his spine:

News coverage of a poll that doesn't favor a political party is not an invitation to meddle in an election through intimidation.I'm urging the National Association of Broadcasters to stand up to this pressure and make sure every station knows what the law actually requires.

If there are good, nuanced ways to make use of generative AI in gaming, as I think there are, that doesn't change the fact that there are nuance-free and terrible ways for advocating for AI in gaming. Tim Sweeney and Epic Games have emerged as helpful foils for the never-AI crowd, given just how cartoonishly brazen Sweeney has been about the use of AI at Epic, as well as the absolute trash vibe-coded launcher the company recently produced. What's going on there almost seems purposefully designed to bolster the arguments from those against any use of AI in gaming.

Epic and Sweeney are not alone in this, it seems. Brandon Stirrup is a business strategy employee at Xbox and he recently went on his own brazen ExTwitter screed that claimed AI use is in far greater employ than anyone is talking about, that games can be easily made by anyone using generative AI tools, and that anyone not using the technology as much as possible doesn't deserve to be in the industry at all.

Stirrup kicked off his thread on October 18 with a simple message saying he'd "vibe coded a bad but improvable game" over two days, although he only "actively" worked on it for about three hours. "I'm telling you I tried it and it's true…anyone can build a game."

After saying that developers and studios who aren't incorporating AI into their strategies are "planning for a fake future," Stirrup went all in.

"Look if I'm fired tomorrow it will be because I told you the truth that nobody wants you to know. You can make a game with a simple prompt and yes you can make custom, tailored games. Everything you want in a game is right there for the taking, for better or worse," he wrote.

This epic online rant went much further than that. Here is a capture of all of it in one image.

There's a ton to hate in there, frankly. Calling gamers stubborn and stupid is one hell of a "business strategy" for Microsoft to be taking here, for instance. I have my own issues with the never-AI crowd, but I have made the point over and over again that their fears are not stupid, not unfounded, and their complaints are not only not to be dismissed, but rather that we need their viewpoint to be taken into account to navigate to a proper type of AI use in the industry. Stirrup does the opposite and dismisses them entirely out of hand. And that is very dumb.

The claim that anyone can make games with a "simple prompt" is also about as obviously untrue a statement as I can think of. Even Stirrup couldn't do so, acknowledging that what he'd managed to produce in his own experimentation was "a bad but improvable game." And that's his own self-assessment. I would hope that a business strategy manager at a gaming company would understand that his opinion of his own product isn't really what matters here. This is what gets so many AI bros in trouble with the public: the constant unsubstantiated claims that AI can do things it simply cannot do. And I imagine that the human artists currently making Xbox games aren't in love with this devaluation of their work, either.

And, finally, stating that anyone who disagrees with him "should not be in the business of being in business" is idiotic. There are many game developers out there making zero or very limited use of AI tools. And they're making great games. There is of course a place in the industry for never-AI developers, just as there is a place in the film industry for directors who make no use of CGI. These are tools, nothing more. Insisting that a painter has no place as an artist if he doesn't use your preferred type of paintbrush makes no sense.

Stirrup deleted his ExTwitter account, claiming he did so because of the volume of responses he got from all of this. I imagine the tone and tenor of those responses had as much to do with it as the volume. And he has only his completely tone-deaf, dismissive screed to blame for any of that. He said aloud he could be fired for what he said. I think he should be.

Not because AI use in gaming should be entirely verboten, but because posting all of that while being employed in a business strategy role seems to prove to me that he's not built for that role.

Age verification proposals are often presented as a simple tradeoff: sacrifice a little privacy to better safeguard children online. But that framing overlooks a more fundamental question. What happens to people who cannot verify their age at all? This is particularly a problem when users are required to prove their age with identity documents.

Age-related restrictions are developing quickly around the world. While they differ radically across jurisdictions, in the past year, we've witnessed a sharp uptick toward mandatory age assurance for social media access but also for other high-risk digital services.

For example:

• In late 2025, Australia became the first country to implement a minimum age requirement for prohibiting children under age 16 from creating or holding social media accounts. 
• India passed the Digital Personal Data Protection (DPDP) Act that includes mandatory verifiable parental consent and has undertaken ongoing discussions for social media restriction.
• The United Kingdom passed the Online Safety Act requiring age verification across a swath of services hosting content considered harmful to users under 18. More than half the states across the U.S. have enacted age verification laws. 
• An EU Commission's expert report recommends a ban on social media access for users under 13. It also recommends mandatory age verification for platforms to ensure that users are 'age appropriate.' 

The Access Problem 

Age verification laws that are predicated on the necessity of users possessing a current passport, driver's license, credit card, or similar documents proving their identity exclude critical sections of populations.  And this problem is global in nature: Millions of people lack government-issued identification or encounter regular challenges to obtaining or updating it. 

Roughly 15 million adult U.S. citizens lack a driver's license, and a further 2.6 million lack any government photo ID; leaving large groups blocked from online content or services. The UK has a similar predicament: proof-of-age checks are often reliant on passports, driver's licenses, or other recognized alternatives, which can pose unique challenges to people without these documents.

Much of the advocacy around age verification discusses the privacy harms of age verification. Yet these measures also threaten something more fundamental: equal access to information and the ability to exercise the right to freedom of expression. 

Unfortunately, age verification systems foster unequal access to information and provide an asymmetric solution to find essential information, build community, and weigh in on public discourse. As governments and companies increasingly require users to prove their age before accessing online services, these individuals risk being excluded from large parts of the internet altogether. 

Effects on Global Majority Countries 

Age verification laws reshape who can speak, who can access information, and who gets excluded from the digital public sphere. In other words, age verification is not a neutral safety measure: it presents a structural barrier to disproportionately exclude certain groups—particularly those in the global majority—and alters the architecture of global online expression. Some of these groups are already marginalized offline, and age verification extends that exclusion into digital spaces.

For instance, 850 million people globally do not have ID. Most of these individuals exist in primarily low and middle income countries in Sub-Saharan Africa and South Asia. The World Bank points out that many are members of marginalized groups and more than half of those lacking access to identity documentation also have children whose births have not been registered. Women are particularly vulnerable and are 8% less likely than men to have an ID. Other vulnerable groups, such as adults in low income countries, are less likely to have an ID when they fall below 25 years, as are those with a primary school education or less or those in rural areas. 

A policy paper from EDRi points out that age verification laws provide quick tech solutions but overlook longstanding structural challenges and undermine the universality of the internet. The analysis finds that age verification laws have serious human rights implications and ironically harm the very individuals they deem to protect. Moreover, these laws depend on the collection of harmful mass data that human rights organizations, including EFF, is fighting against—and has for decades. 

For example:

• In Morocco, a push to restrict children's access would ban under-13s from creating accounts on gaming platforms. This could potentially create challenges for those with IDs that have incomplete information concerning the year of birth. In addition, this push would also impact those who attempt to leave Morocco.
• In Egypt, the lack of ID cards has created challenges for minority groups such as the Baha'i.
• Stateless individuals would also struggle under age verification laws. The Rohingya, the world's largest population of stateless people, often lack IDs. 
• Kuwait's Bidoon population also lacks proper identification materials. According to Amnesty International, this leads many to rely on standard civil identity cards, which can be restrictive.  
• Nigeria launched a national program to provide ID cards to its population in 2007. Since then, 64.4 million have registered, but that represents just over 30% of the national population.
• The Kafala system, a practice in several countries across the Middle East, also introduces challenges for age verification laws. Under this legal framework, migrant workers' legal residency and employment status are bound to a specific employer. This potentially leaves stateless persons or migrant workers vulnerable to forced labor and restricted movements. 

No ID, No Access

Age verification laws are less about confirming the age of a user and more about creating barriers to online participation that many people cannot reliably scale. The net result is reduced access, more data collection, and an increased chance of unequal or mistaken exclusion from accessing information online. 

Nobody doubts the importance of protecting children online. While proponents assert that age verification laws protect minors from harmful material, online harassment, and digital addiction, these laws are not the solution. They subvert fundamental freedom of expression rights and pose significant privacy risks.  

EFF has long warned against age-gating the internet. Age verification technology itself is often inaccurate and privacy-invasive, and as more countries considering implementing ID-based checks, the risks move beyond censorship and surveillance toward excluding some people entirely. That's why we're working with groups in the U.S. and around the world to push back against these laws, and why we hope you join our effort.

Republished from the EFF's Deeplinks blog.

It wasn't until after immigration officers started murdering US citizens that the Trump administration finally agreed to do anything about it. First, Trump turfed Kristi Noem, setting her up with a "special envoy" dead end job (which has also now come to an end). Then it elevated Markwayne Mullin, who first insisted his priority was to tone down things a bit to keep ICE from generating negative press on a daily basis.

Pressure was applied to outfit ICE officers with body cameras. Hilariously, DHS officials claimed there simply wasn't any money to do this, despite ICE now commanding the most money of any federal law enforcement agency. More money was given to ICE to handle this fake budgetary shortfall, which then led to a lot of things, none of which did much to rein in ICE's tendency to abuse its power, violate rights, and engage in violence.

First, the rollout was slow-walked. The deadline for deployment kept being pushed back. When full deployment finally arrived, it was pushed back even further by vague claims about delays in training. Then the DHS made it clear body cam footage would be almost inaccessible by the public. Finally, the cameras were being worn by officers but somehow never seemed to be rolling when officers violated rights, abused their power, or engaged in violence.

The Government Accountablity Office was asked to look into ICE's body cam program. The resulting report [PDF] noted that while this was better than not having cameras, it left a lot to be desired.

The report, requested by congressional Democrats, comes as they have fumed about slow deployment of the cameras as well as ICE officers' failure to activate them in a string of shootings, including during multiple fatal interactions. ICE officers have killed six people since President Trump took office.

"ICE has not implemented controls to help ensure compliance with activating cameras when required by policy," GAO concluded in its report.

First of all, while ICE has promised to outfit all field officers with cameras, it has continued to delay the rollout because it is continually onboarding new hires. This is a problem that's never going to go away — not when the administration is constantly demanding more officers and more arrests and the agency isn't able (or willing) to throw more money and personnel at training programs to ensure new hires are both trained and outfitted with these mandatory recording devices. According to what's stated in the GAO's report, ICE has decided the best stop-gap solution is to… just stop buying more body cameras.

[A]s of June 2026 ICE has not updated its plans to reflect changes stemming from an increase in the total number of cameras scheduled for deployment and schedule delays caused by the pause in further camera deployment. Further, ICE, like CBP, is at least one year behind its 2024 plan.

Additionally, ICE's overall deployment needs have increased. Specifically, according to ICE program officials, the component intends to provide cameras to all personnel, including planned future hires. In July 2025, ICE initiated a hiring surge of about 12,000 additional personnel. However, the number of cameras scheduled for deployment documented in its 2024 plan has remained relatively stable as of April 2026.

While both CBP and ICE have policies that state officers must start recording at the beginning of "encounters," only CBP seems interested in actually making this happen. It does have some auto-record options deployed. The GAO recommended ICE do the same thing. And despite both agencies being housed by the DHS, ICE has affirmatively stated that it won't implement auto-activation features in cameras worn by ICE officers.

ICE disagrees that it is necessary to implement additional controls or procure the accessories required for automatic activation, which triggers upon the drawing of a firearm or TASER. Specifically, automated activation presents significant operational risks, including inadvertent or false activations during routine movements.

ICE then goes on to list the possible "harms" of auto-activation, which include recording discussions with "informants or victims," as well as the "inadvertent" documentation of First Amendment activities. This might mean something if (1) ICE didn't control the recorded footage, (2) allow officers to record people engaged in protected First Amendment activities, and (3) ICE officers didn't routinely leave their cameras off when engaging in violent acts.

In the end, this is nothing more than ICE telling everyone — the people it serves and the people it answers to — that the body cameras are nothing more than expensive lip service. These devices aren't going to be used to hold officers accountable. If they're used for anything at all, it will be to find more charges to throw at arrestees or selectively released to support the government's preferred narratives.

The big Meta settlement with a bunch of states got most of the headlines, but state AGs have gone all in on suing tech companies of late, and it shows no signs of slowing down. Last week things went ballistic when the judge in New Mexico's case (New Mexico being one of the few states that refused to join the others in settling) ordered Meta to pay a $25 billion penalty. As Daphne Keller points out, a rough back-of-the-envelope calculation shows that $15 billion of that judgment covers the same claims that all the other states together settled for just $459 million (though Meta has said it will fight the ruling).

As we discussed earlier, TikTok also settled with Alabama in a separate case. But the cases just keep coming. Rhode Island just sued Meta (again), this time claiming it allowed scam ads to run rampant on the site. And a group of state AGs (mostly using the same outside law firm) sued OpenAI separately (same basic lawsuits filed individually in each court). TikTok, Snap, YouTube, all face additional similar lawsuits as well.

Many people are cheering these lawsuits on, and you can understand why. These are some of the most hated companies on the planet right now. No one trusts the big Silicon Valley companies and their broligarch CEOs who seemed willing and eager to fall in line (literally) behind the fascism of Donald Trump. Not to mention the enshittification. Oh so much enshittification. And then add the relentless attempts to force AI tools into places people clearly don't want them.

People were understandably and reasonably fed up. And they wanted blood. Massive verdicts deliver it.

But as I've warned, we shouldn't be too quick to celebrate these kinds of things. Indeed, we may quickly be learning exactly why I was so concerned with those early verdicts against Meta earlier this year. As I noted at the time, some of those cases were the kinds of lawsuits that Section 230 was supposed to protect against — and for good reason. Other cases (like the latest New Mexico ruling) do involve statements by the company (thus they're not Section 230 issues) but are still equally problematic, as we'll discuss below.

First, most of the product liability, negligent design, and kid safety cases really shouldn't even exist under Section 230. Effectively all of them are predicated on the fact that some people do bad things, some (not all) of which were loosely connected to the internet services that are being sued, under the theory that the internet companies should have prevented the bad things from ever happening. But the entire premise of Section 230 was that if you hold internet platforms liable for every bad thing that happens on them, they will massively crack down on speech online, to avoid facing liability. It turns the internet from being a global communications network… into a bland "broadcast" tool, where only approved content gets shared.

It also makes it almost impossible to run an online platform that allows third party speech unless you have massive resources and a building full of lawyers. In other words, it means that only the largest internet companies can survive, while making it way harder for there to be any new entrants in the market.

There's a reason that State Attorneys General spent many, many years pushing Congress to amend Section 230 to carve out the enforcement of state laws. They never actually got that. But they may have gotten something better. A bunch of judges who decided that state AGs and the courts can simply ignore Section 230 and hold internet companies liable for things their users do, so long as you claim the bad things were due to "product design" or "negligence." The Ninth Circuit recently eviscerated the key part of the law that was supposed to stop exactly these kinds of cases.

It's the "get around 230" trick that collapsed the entire law.

As for the New Mexico case (and some of the new Rhode Island case), they are, in theory, predicated on statements by the company. But when I look at the New Mexico ruling (even setting aside the size of the judgment), I can't figure out how it makes any sense. The jury verdict is based on the claim that public statements by Meta and some of its executives were "willfully deceptive." And, as someone who doesn't trust Zuckerberg or Meta at all — and who has called out some of his willfully deceptive statements in the past — I was primed to side with the jury.

But look at the "statements" in question. I'm having trouble seeing how any of them are actually deceptive in any meaningful way. I'm not going to post them all, but they all seem mostly true, even if the public believes otherwise. Take for example:

"We don't sell your data and we don't tell advertisers who you are. What we do is allow advertisers to reach people interested in their products."

This statement (by Sheryl Sandberg in this case) is… accurate. I know that, broadly, people often claim that "big tech sells your data" but that's never actually been true. Their business model has always been based on selling advertisers access to your feed based on your data. That's a very different thing. What the company does is allow advertisers to target you based on the data Meta has about you, but the advertiser never gets or sees any identifying information from you or data about you. There are data brokers out there that do sell that kind of data (and those should be wiped off the earth, as far as I'm concerned), but that's never been any of the big tech companies' business. Indeed, one of the statements the jury found deceptive was Mark Zuckerberg accurately describing this dynamic:

"There are other internet companies or data brokers or folks that might try to track and sell data, but we don't buy and sell."

That's an accurate statement! But the jury found that this one statement alone generated 1,386,648 violations of the law! What?!? How that and all the other "violations" (almost 44 million of them!) came about is also highly questionable. It assumes that each New Mexico resident who might have come across one of the statements counts as an actual violation under the law:

It assigned 2.1 million violations to each of 11 statements that appeared in major news outlets and 1,386,648 to each of 15 statements made through company channels, including its newsroom, executives' Facebook posts, earnings calls and a developer conference, for a total of 43,899,720.

According to reports, 2.1 million was the population of New Mexico in 2020. 1.4 million was the number of New Mexico residents with active Facebook accounts at the time. So basically the jury claimed that each person who could have seen one of those statements because they appeared in a major news outlet… equals a violation of the law. And for the statements made on Facebook/Meta properties themselves, they assume every person with an account could have seen it, making it another violation. That calculation seems… bizarre. Lawyers are already saying they doubt it will hold up on appeal.

Even worse, Meta execs also got dinged for general statements about how they try to take down misinformation and don't profit from misinformation:

"Well, if we see harmful misinformation on the platform, then we take it down. It's against our policy. So the 18 million number that I shared is the number of pieces of content that we've seen on the platform that we take down. Now, do we catch everything? Of course there are mistakes that we make or areas where we need to improve, but that's the best number that we have in terms of what we've seen and what our systems have been able to detect."

That's Zuck again, and that statement is also… pretty accurate? They certainly do miss some harmful content, but that is the nature of content moderation, especially when you have billions of users around the globe. I don't see how that claim can be deceptive.

In court, New Mexico apparently argued that this was deceptive because of Meta's Xcheck program. But as we were just discussing last week, there was a perfectly valid reason for Xcheck. It was the program that was put in place to deal with high profile accounts. Those were the accounts where Meta recognized that if they made a mistake, and took down those accounts incorrectly, it would make the company look really bad. So they created a system where no action would be taken on those accounts without a secondary human review. Given the situation, this sounds reasonable. It's what most companies would do facing similar challenges (and, it is, in fact, what most companies have done).

But when Frances Haugen released the details of this program, it was reported on (misleadingly) as "Meta allowing the elite to violate its rules." But that's not what the program was for. It was to make sure there was additional human review, even if that review sometimes took too long.

But New Mexico presented it to the jury as Meta claiming to take down harmful misinformation while secretly using Xcheck to let big accounts spread it, and the jury (and now the judge) punished Meta accordingly.

I'm honestly not sure how any internet company avoids facing a similar risk down the road. When an AG can take broadly true statements, strip them of context, and call them deceptive (when the only deceptive thing is the AG's presentation), how can any company survive such lawsuits?

And, with the chance to get a $25 billion award from it, these lawsuits are going to be everywhere. It could even be worse, given that New Mexico actually asked for even more, and the judge said he could have ordered a penalty over $200 billion. And, again, these are for basically true statements! How do you run an internet company with that kind of threat of liability over your head?

Even if you look at the strongest elements of New Mexico's case against Meta, I'm left a bit baffled. The key part of the case resolved around the claim that "You have control over how your information is shared," which was a statement from back in 2010. The court in New Mexico claimed that was deceptive… because of the FTC's $5 billion fine of Meta, over the Cambridge Analytica scandal. That fine hinged on the same claim, with the FTC arguing that it was misleading by Meta to claim that because anyone could also share their friend graph, and since that graph includes information about other people (namely that you are connected to others), they argued that revealed private info.

But, as we noted at the time, that doesn't seem like that big a deal? Knowing who is in your friend graph from a social media site doesn't really seem like violating privacy. Friend graphs are… just how social media works. There's also no evidence that anyone was directly "harmed" from the ability to access a friend graph this way.

Finally: the FTC fined Facebook for this already, and did so at $5 billion for the entire country. Why does anyone think that New Mexico alone deserves 5X that amount?

Combined with the judgments and the settlements, these cases are going to keep piling up, and they're going to make the internet worse in almost every way. If you thought enshittification was bad before, wait until politicians and judges are the true product managers for the internet, driven by headline seeking and profit maximizing Attorneys General who will view suing tech companies as a way to fill state coffers.

People can celebrate the fact that big internet companies we all hate are losing cases and having to pay, but the end result won't be a better internet. It will be one driven by politicians looking to shake down tech companies any time anything "bad" happens online. That's going to lead to way more restrictions on the internet. Way more identity verification to do anything. Way less innovation (and new offerings will have to go through an exhaustive legal review — not to make them better, but to make them less likely to end up in court). It's also going to lead to way less transparency. The New Mexico verdict means that even accurately describing how your site works can lead to a massive verdict against you, so companies are going to tell execs to shut up completely. This will also mean very few new competitors — except those with massive bank accounts. The broligarchs will be able to afford the liability risks of new services. Everyone else will be priced out.

This was the world that Ron Wyden and Chris Cox sought to avoid when they authored Section 230 30 years ago. But it's here now — even without the repeal of Section 230 — and we're going to see just how bad that is for the internet.

Browse, plan, and prep every trip from one library. You get 30 Lonely Planet travel eBooks in one bundle, packed with bucket-list destinations, must-see sights, and insider tips for your next trip. Instead of buying guides one at a time, you build a full travel library in a single purchase. Your guides live in one digital library that you can open on your choice of laptop, tablet, or phone. You have no heavy stack of books to pack. You can pull up the right guide whether you're daydreaming at home or waiting at the gate. Maybe you're mapping out your first international trip, or maybe you're adding to a wish list you've kept for years. Either way, you get 30 starting points in one place. Pick a destination, open the guide, and start building the trip you've been putting off. It's on sale for $30.

Note: The Techdirt Deals Store is powered and curated by StackSocial. A portion of all sales from Techdirt Deals helps support Techdirt. The products featured do not reflect endorsements by our editorial team.

The thing about being in power is that there's not much room left between the floor and the ceiling. The administration can fall, but only so far, because it's still the administration. Conversely, it can only rise so far. The system of checks and balances — even when broken — limits upward trajectories. Bureaucracy generates plenty of friction on its own, which means even when everything aligns perfectly, things still move slowly and rarely reach the heights promised or projected.

Trump's anti-voter operations are banging their collective heads against the wall. The main man continues to insist millions of illegally registered voters exist. The operational side of this obvious lie continues to struggle to convert this hyperbole to reality. Sure, it tries to lay the groundwork for the lies, but the moment anyone pushes back, the administration has been forced to admit it doesn't actually have any facts to back up its accusations.

This attempt to disenfranchise voters, cast doubt on any future election results, and discourage people who might not vote for GOP candidates from heading to the polls is backfiring spectacularly. In California — a state Trump often portrays as being completely overrun by dangerous migrant criminals — the DOJ and DHS have delivered next to nothing. The DOJ brags about its 28 arrests of "illegal" voters, but those have happened in a state that has more than 23 million registered voters.

The administration is apparently tiring of having to do all the heavy lifting when it comes to propping up its own false narratives. Under the threat of further legal action by the DOJ, Trump is now demanding California engage in a witch hunt of its own.

The Department of Justice sent letters on Friday to local election officials across California demanding that they help identify potential noncitizens on their voter rolls and set aside their ballots.

The letter — sent to every county in the state, according to a spokesman for the U.S. attorney's office in Los Angeles — also said local officials would "risk significant federal scrutiny" if they failed to comply with the Justice Department's demands.

California is, of course, under no legal obligation to do this. Its obligations to ensure the legitimacy of voters are already subject to local laws and other efforts it already engages in. This bizarre demand doesn't have any legal weight behind it. But it does contain a vague threat that the DOJ could easily turn into a reality by directly interfering with the election process during the mid-terms.

It certainly seems like that's what the administration will do, whether or not California officials agree to do Trump's dirty work for him.

Last week, the Trump administration sued California, a heavily Democratic state, seeking to temporarily pre-empt the counting of more than 20,000 mail ballots sent to people it said were suspected of registering to vote for the Nov. 3 election, even though they were not U.S. citizens. 

That was the opening shot. The 28 arrests the DHS touted are the second wave. This threat — which comes with a 48-hour compliance expectation — is the latest. There will undoubtedly be more, even if the administration loses its attempt to block more than 20,000 ballots in court.

The counties receiving these lists have been given less than two days to validate hundreds to thousands of issued ballots. The government's demands are already ridiculous, as it expects state officials to prove negatives as well as positives. But they're also ridiculous in a much, much stupider way: it appears the DOJ hadn't bothered to vet its own list of suspected "illegal" ballots before sending out its mass email to California election officials.

In rural Inyo County, along the Nevada border, the registrar of voters said the list she was sent included just two people, and neither of them had been issued ballots. Danielle Sexton, the registrar, said her records showed they had never voted anywhere in California.

And in Mendocino County, a rugged stretch on the north coast, the registrar said the list she received was completely wrong.

"It's not our voters that were listed," said Katrina Bartolomie, the Mendocino clerk-recorder. "It's from another county."

Par for the course for the Trump administration, where high-level officials view themselves as snipers, but always opt for birdshot and sawed-offs during the loadout process. But they know who they work for, so it probably doesn't matter. We should absolutely expect the administration to disrupt the election process and engage in protracted legal battles everywhere the GOP loses a seat. We should also expect the DOJ, DHS, and others to contradict the public assurances they've offered and send feds to polling places to suppress voting in contested precincts.

What's on display here is just the groundwork being laid for future bullshit challenges of the legitimacy of an election that, at this point, looks to give the Democratic Party the control of the House.