SAP has taken meaningful steps in opening its data to bidrectional sharing across the enteprise, and in decoupling the advanced analytics needed for AI from its ERP upgrade path.

During a technical conference last week, the German software giant adopted native support for Apache Iceberg, the open table format, and extended its partnership with Databricks to bring data from legacy ERP systems into its modern cloud analytics platform.

"By putting [a] stake in the ground with Iceberg, we now have SAP moving from a very closed application storage model to an open lakehouse model," said Aaron Rosenbaum, Gartner senior director analyst, referring to the concept which brings together data lakes and warehouses.

Also important was the bridge between modern analytics in the cloud and on-prem or hosted legacy systems such as ECC, the ERP system broadly adopted by global companies. "SAP made a pretty big move in saying we're going to open this up, not just to S/4HANA Cloud [the latest ERP system], but ECC users. That coupling — leveraging of AI, leveraging a lakehouse for AI to the ERP upgrade lifecycle — was a problem," said Rosenbaum.

In May this year, SAP bought Dremio, a lakehouse platform built around Apache Iceberg, ending its previous preference for Databricks' Delta Sharing to bring together SAP data products with existing lakehouses bi-directionally.

At SAP's TedEd conference last week, SAP said its Business Data Cloud (BDC) - the "trusted data foundation for SAP Business AI Platform" - would now support native Apache Iceberg and a universal catalog powered by Apache Polaris, "providing an open and governed data foundation for users and [AI agent platform] Joule to consume data."

Together with SAP BDC Connect, the new tech policy will support platform interoperability with "zero-copy data sharing" across platforms from Databricks, Snowflake, Google, AWS, and Microsoft. The system would also continue to support Delta Lake, the alternative open table format from Databricks, SAP said.

With the integration of Dremio's SQL query engine, BDC users will also get access to relational databases including SQL Server, MongoDB, Oracle, and PostgreSQL without moving data, the vendor said.

During the same conference, Databricks - the lakehouse company born out of an Apache Spark platform - extended its partnership with SAP based on the previously announced SAP Databricks, which embeds the Databricks Data + AI Platform in BDC.

That "expanded scope" promises to make data from SAP applications available for consumption in the SAP Business Data Cloud engine of choice, including SAP Databricks. The vendor said that a manufacturer running finance and supply chain on SAP ECC on-premises, for example, could use SAP Application Data Foundation to bring historic order data into SAP Databricks, combining it with third-party data, and allowing data engineers to build demand-forecasting models on SAP's business semantics, without custom extraction pipelines. They can keep that work as they move to the cloud, Databricks suggested.

This being 2026, the hoped-for output from all of this data engineering is not just business insight, but automated action, by way of AI agents. SAP has its Joule AI assistant and agentic platform, while Databricks has Genie, and the two companies have promised to bring them together.

The increasing openness is good for users who will get more options in plotting a strategy to bring together enterprise data, leaving them pondering over where to put the business context layer, which vendors will no doubt fight to control, Rosenbaum said.

"You have the comment that they're going to bring Joule and Genie together in the future. This is pointing to a struggle that's going on throughout enterprises and throughout all the vendors about the context layer. I don't necessarily see that there's one winner. You do have different points of context within the transaction system and across the whole business, and they have been points of tension.

"Lack of single vendor lock-in is important. It's something that all the customers are struggling with: how do we build these context layers and deliver sort of consistent answers?"

SAP's shift in the direction of openness is likely to help users in this struggle, whether they choose to build the context layer within the ERP vendor's data platform using its semantic layer, or outside of it using tools from other vendors - there are plenty of companies offering help for those opening their wallets.

Nonetheless, whichever path users choose, a lot of work lies ahead. ®

Royal London plans to cut hundreds of IT jobs and shift work to India under an outsourcing deal with IBM, according to sources, with redundancies expected by the end of December.

Sources said Royal London's chief technology officer, James Sandell, told IT staff in September that the insurance and pensions giant was partnering with IBM to outsource a substantial portion of its technology operations. A consultation process was expected to follow.

A Royal London spokesperson told The Register: "Royal London has appointed IBM as its strategic technology infrastructure partner as it continues to develop its technology strategy. The partnership will give Royal London access to IBM's specialist capability, scale and experience. Royal London will remain accountable for service delivery and regulatory requirements and continues to focus on ensuring continuity and quality of service for customers."

We asked Royal London how many UK IT roles it would retain and whether IBM would move work to India.

The exact number of jobs at risk remains unclear, although one Glassdoor review posted on October 6 claims that hundreds of positions in the company's technology operations division are being eliminated.

"A vast amount of knowledge is being lost by the business, a lot that is not yet documented due to constant years of recruitment freeze and workload volume," they wrote.

The reviewer also questioned whether management had grasped just how much work would be involved in handing operations over to IBM, particularly with already stretched teams being cut to a fraction of their former size.

Royal London replied to the Glassdoor review, acknowledging staff concerns about the changes, though it didn't comment on the claims about outsourcing and redundancies.

"We recognise that as Royal London is going through a period of transformation, this has raised questions and concerns for some colleagues," the company wrote. "We have established channels for colleagues to raise these, including through People Leaders and Colleague Representatives, and I encourage colleagues to use these routes to discuss their concerns."

Other October reviews questioned the handling of the outsourcing and Royal London's commitment to its mutual values.

An October 12 reviewer identifying themselves as an IT infrastructure employee said the company had historically offered good career prospects, before criticizing its new direction.

"Sudden outsourcing goes against all Royal London values," the reviewer wrote, adding that employees were questioning the company's direction amid reports of a hiring freeze and further potential cuts.

Another review, posted on October 20, suggested the IBM deal was the latest in a series of outsourcing decisions following an earlier move to transfer servicing jobs to South Africa.

"A mutual company but not mutual values when it comes to things like this," the reviewer claimed.

The planned cuts follow rising profits. Royal London reported adjusted group operating profit of £187 million in the first half of 2026, up 13 percent year-on-year.

Royal London has been keen to trumpet the benefits of its mutual structure, under which it is owned by customers rather than shareholders. Reporting the results in August, chief executive Barry O'Dwyer highlighted the £199 million distributed to eligible customers through ProfitShare in April.

"As a customer-owned business, when we do well, our customers share in that success," O'Dwyer said.

For employees facing redundancy before Christmas, the benefits of that success may be somewhat harder to appreciate.

The Reg has asked IBM to comment. ®

Google Cloud has revealed that the cause of the October 20 outage in its europe-west-2-b availability zone was a 'misconfiguration' of its systems that meant its emergency generators didn't start.

The outage at the London facility caused service interruptions at outfits including Lloyds, HSBC, Royal Mail, and Revolut.

In a November 4 update to its service status page, Google explained what went wrong.

"The site relies on a 2N utility power architecture — where the facility has complete duplicate power paths for every critical load — backed by a central high-voltage generator bank managed by a dual-redundant Programmable Logic Controller (PLC)," Google revealed.

The facility gets its juice from two independent utility feeders, which Google calls "A" and "B". There's also a "standby utility connection configured for manual campus-wide transfer during maintenance operations or emergency events."

Google says the root cause of the incident was failure in an external underground cable through which its power supplier brings 'leccy to its bit barns.

That cable connects to primary feeder A and it broke at a time when feeder B wasn't working.

"Circuit breaker feeder B was physically opened to allow for corrective repairs following issues identified during a quarterly generator load test on 14 Oct 2026," Google admitted. "Consequently, primary utility feed A was supporting the total critical load while the feed B breaker remained in an open position to allow repairs to be conducted on the B feeder pathway."

That shouldn't be a problem for a top-tier operator like Google, which knows it needs to install backup sources of electricity like uninterruptible power supplies (UPS) and on-site generators.

Indeed, europe-west-2-b employs UPSes that Google says can keep things up and running for five minutes if the availability zone loses its connection to the grid.

The UPSes worked, but a "voltage-sensing design flaw and PLC misconfiguration prevented the facility's generator plant from receiving an automatic start signal to assume the critical load."

Google's UPSes therefore ran out of juice, but its generators didn't start. The rest is history.

The search giant has figured out that its PLCs weren't set up to handle the conditions it faced on this fateful day.

"PLC logic indicated that a healthy utility source was available, even though the primary and secondary distribution feeds were de-energized, and did not issue a start command to the generator plant," the status update reveals.

Once Google was able to reconnect to its power utility, it reprogrammed the PLC so it wouldn't make the same mistake again.

The Big G has since made many changes to the electrical setup at europe-west-2-b and says it's tested them and that they all work.

It's also given itself two action items, one of which it describes as a "Communications Enhancement" that will see it "bolster notification speed by developing an automated alert-to-communication mechanism for location-impacting incidents."

The other is to "expand platform-wide testing outcomes for infrastructure-level events."

Once Google gets that done, perhaps it will be able to handle outages better - and if not, at least advise customers about its troubles more quickly. ®

VMware appears to have made a concession to European regulators by extending the life of its low-end vSphere Standard and Enterprise Plus products.

The virtualization pioneer quietly announced the extension in the November 4 blog post in which it announced a new low-end product called vSphere Express, which The Register reported yesterday.

We based our report about vSphere Express on an interview, and on material provided by VMware - which sent a version of the blog post linked to above.

The version we saw did not include the following paragraph, the contents of which were not mentioned in our interview:

"Furthermore, for customers in countries in the European Economic Area, vSphere Standard version 8 and vSphere Enterprise Plus version 8 will remain available for sale until October 31, 2027, and will have Extended Support included in any active subscriptions to those offerings until the earlier of the applicable subscription end date or October 31, 2028."

The latter date is more than a year past the October 11, 2027, end-of-general-support date for the more capable vSphere 8 and Cloud Foundation 5.x products.

"This has the potential to be a positive development for organisations within the European Economic Area," independent analyst Michael Warrilow told The Register. "It could provide them up to an extra 12 months to transition vSphere 8 to VCF 9 and/or migrate workloads to another platform."

Warrilow thinks European users should seek "absolute clarity" from VMware about the specifics of its offer.

"What level of extended support will be provided and what longer-term commitments will need to be made? I cannot find any of that detail," he wrote.

A reminder: VMware has promised in the past to make patches available for users who choose to persist with perpetual licenses instead of moving to its preferred subscriptions. The Broadcom business unit later paused patch downloads for months, claiming it needed to adjust its website to make them possible.

Warrilow wonders why VMware will keep selling vSphere Standard only in Europe and suggested "regulatory muscle" is the reason.

Regulators around the world cleared Broadcom's acquisition of VMware. In Europe, authorities have since revisited the matter.

Lobby group Cloud Infrastructure Services Providers in Europe (CISPE) has succeeded in having the European Commission open an antitrust investigation into whether VMware's channel changes inappropriately shaped the cloud infrastructure provider market. The Register understands CISPE feels regulators are receptive to its arguments.

In the Netherlands, local courts enforced a VMware support contract of the sort that Broadcom has tried to walk away from at other users.

Warrilow wonders if VMware might now change its policy regarding vSphere Standard in other countries.

"The subsequent question is how can organisations in other regions get the same concessions from Broadcom? I would encourage them to amplify their voice by rallying industry associations and user groups to the cause whilst lobbying the regulators in their jurisdiction."

Overall, however, he thinks VMware's decision isn't enormously impactful.

"Ultimately, all it does is buy a little more time to upgrade or migrate," he wrote.

It may not even do that, as The Register often heard that VMware seldom issued quotes for vSphere Standard or Enterprise Plus.

Paul Turner, chief product officer for VMware's Cloud Foundation Division, told The Register that inappropriate incentives for channel partners caused that situation, and that VMware later fixed it.

Continental users will be hoping VMware got that effort right. ®

When The Register writes about professional certifications offered by vendors, we often receive comments arguing that such credentials are nowhere near as valuable as real-world experience. Now, Microsoft has acknowledged there's some merit in that argument.

Redmond revealed its stance last week in a post from general manager for content and insights Cameron Percy, who observed: "For years, people have asked for a simpler way to prove the skills they use in the flow of work, without stepping away to prepare for another test."

Microsoft's response is the "Pro Badge," a new form of credential that Percy wrote "works alongside Microsoft Certifications."

"Certifications show that you can solve problems in a role or field. Pro Badge shows how you use a specific product in your everyday work. Together, they give a more complete picture of what you can do."

"Instead of capturing skills on a single test day, this approach keeps pace with your work and the products you use," Percy added. "It gives you a current, verified signal of what you can do, at no cost and without a separate exam or study guide."

The first Pro Badge is for GitHub Copilot. To earn it, users need only visit learn.github.com and sign in with their GitHub account.

Going through those steps fires up a "proficiency engine" that Microsoft says "gets to work analyzing relevant product usage signals and evaluating them against a defined proficiency framework." Percy's post says it "systematically uses evidence from everyday product use to calculate and automatically update your proficiency level."

Microsoft offers five levels of proficiency: beginner, intermediate, proficient, advanced, and expert. To earn a Pro Badge, users must reach the proficient level.

A beginner has had "first contact" with GitHub and "understands what Copilot is, where it can be used, and how it might fit into everyday development workflows."

An intermediate user is "familiar with web-based collaboration features, core IDE features, model management, agent configuration, customization options, and techniques for managing context."

To score the Pro Badge, users will need to demonstrate "practical ability to work on projects" and use GitHub "across typical scenarios, demonstrated through consistent hands-on usage across multiple repositories and extended periods."

"A Proficient user can use many web and IDE features, select appropriate models, and apply customizations to tailor Copilot for specific project needs. They have experimented with extended features and are aware of extensibility/integration options."

At the time of writing, Redmond hasn't defined expert or advanced proficiency. It has published more criteria for the first three levels here.

Microsoft has kindly integrated GitHub's learning site with LinkedIn, which it owns, so users can add Pro Badges to their profiles.

Percy's post suggests Microsoft has other Pro Badges in the works.

Elsewhere, Microsoft makes it plain that Pro Badges won't replace its other credentials.

"GitHub Verified Proficiency is part of Microsoft Pro Badge, a new credential type designed to recognize skills as they are applied in the flow of work," explains a page offering detail of the new cred. "It does not replace certifications or Microsoft Applied Skills. It adds a third signal: how skills show up through everyday product use."

The Register welcomes comment on this new approach to professional credentials. ®

Google-owned autonomous cab company Waymo has revealed it will stop allowing Uber users to book its robo-cabs, hours before the OG rideshare giant made a case for all autonomous vehicle fleet operators to use its platform instead of creating their own.

This slow-motion fender bender played out across Wednesday, starting with Waymo Co-CEO Tekedra Mawakana telling Bloomberg that her company would end a partnership that saw it work exclusively with Uber in Austin and Atlanta.

Mawakana said Waymo wasn't happy about the fact that when users of Uber's app requested a ride in one of its robo-cabs, they weren't guaranteed to get one.

News of that break-up came on the same day that Uber announced its quarterly results. As is common on such occasions, the company published remarks [PDF] its execs intend to make during an earnings call.

CEO Dara Khosrowshahi used those remarks to promote the idea that Uber is the natural home for robotaxi fleet operators, because it already delivers around 30 percent higher trips per vehicle per day compared to apps run by autonomous cab companies. He said that higher utilization rate, plus use of shared charging and maintenance infrastructure, can deliver "higher fleet profitability even after paying Uber's marketplace fee."

"As autonomous technology becomes more widely available, we believe this ability to commercialize vehicles profitably will become an increasingly important source of competitive advantage and value capture."

The two events are deliciously ironic, because Uber is arguing the scale of its platform makes it the ideal host for third-party players - just as Google does with its search, app store, and YouTube services. Left unsaid is that the two companies hold concerningly dominant positions in their respective markets.

Khosrowshahi also outlined more ways Uber intends to strengthen its position, including by putting around 2,000 robo-cabs on the road in 2027, before "a much larger commercial ramp beginning in 2028." That plan means Uber will run some cars itself and rely on third-party robo-cab operators who allow bookings on its app.

"Importantly, building a broad ecosystem does not mean spreading our resources evenly," the CEO explained. "We intend to maintain an open platform while concentrating capital, integration resources, and hands-on deployment support behind a smaller group of partners with the clearest paths to commercial scale and attractive economics at maturity."

Or in other words, as Uber uses more robo-cabs, it will try to maximize profits.

Uber users booked $59.8 billion worth of services in its third quarter, up 20 percent year-over-year. The company recognized $14.75 billion of that as revenue, a ten percent improvement.

Rideshare accounted for $29.5 billion of bookings and $7.5 billion of revenue. Delivery services brought in $5.5 billion of revenue on $28.5 billion of bookings.

No wonder Khosrowshahi wants more robo-cabs: they'll boost Uber's more lucrative business.

The company might need that boost because Australia's High Court yesterday decided [PDF] that the company must pay AU$81 million ($57 million) in State payroll taxes because the fees paid to rideshare drivers are effectively wages. While the case considered Uber's activities in one of Australia's six States, all of them charge payroll tax. The case is one of many around the world in which courts have found gig workers should be treated as employees for legal and tax purposes. The High Court is Australia's highest judicial authority and like-minded jurisdictions around the world are aware of its decisions, potentially making more trouble for Uber further down the road. ®

Meta has acquired AI coding startup Conductor, signaling interest in applying AI to software engineering tasks.

Charlie Holtz, the startup's co-founder and CEO, made the announcement in a blog post. He also revealed that the company's hosted Conductor Cloud, a sandbox for hosted agents, will go read-only on November 13 and shut down on November 20.

Conductor, the eponymous company's macOS app for running multiple coding agents simultaneously in isolated copies of Git projects, will continue working for the next few months, and the biz expects to provide a way to transfer Conductor Cloud workspaces to local machines.

But in early 2027, Conductor will no longer receive new models or critical bug fixes.

A Meta spokesperson confirmed that the acquisition will strengthen the company's developer-oriented AI capabilities but declined to elaborate further. Perhaps Conductor will play nicely with Muse Code, Meta's recently introduced coding agent.

Having spent heavily on strengthening its AI research lab, Meta has moderated its appetite for big acquisitions since April when China blocked its $2 billion offer for Manus AI, a developer of software agents.

Meta's other acquisitions this year, Moltbook, Assured Robot Intelligence, and Stilla, are also AI-oriented but more modest in scope.

The Conductor deal comes without an announced price, and follows in the footsteps of Meta regaining credibility as a model provider: Llama 4 disappointed but Meta's recent Muse Wonder and Spark models rank respectably among the top LLMs at the moment.

In Meta's most recent conference call for investors [PDF], CEO Mark Zuckerberg said his company aims to provide AI agents that non-technical people can deploy for business tasks.

"What we see is a lot of people are actually using Muse today to help run especially small businesses," he said. "And we think that there's both an opportunity to help create an agent that people use to really enable them to automate and do a lot of the parts of spinning up a small business that are just kind of difficult and time-consuming but aren't the core thing that a person is trying to do."

Meta's AI strategy, as Zuckerberg describes it, is focused on collecting transaction fees from businesses for enabling agentic commerce, selling advanced AI capabilities to developers in the form of model access or tokens, charging for Muse to automate business tasks, and applying AI to personalize and improve results from ads.

It's a good time to run an AI-related startup, as leading model companies seek to add value beyond offering AI models. Models themselves have become commoditized to the point that only a few leading model makers can get away with charging a premium. In an effort to maintain that lead, top model makers have been purchasing innovation (and talent) while they are still flush with investor funds.

On Wednesday, OpenAI said it plans to acquire Pierre Computer Company, maker of distributed storage service Code Storage⁠ and open-source developer tools. And a few months earlier, the ChatGPT maker snapped up Ona, which makes cloud development environments for agents.

Anthropic's most recent developer tools deal involved Stainless back in May. Its purchase of Opaque Systems, which offers confidential computing technology, could also be squeezed into the developer tool category.

At some point, when it becomes clearer what kinds of AI services people will pay for, startup acquisitions may become more focused. But for the moment, with software development being the most successful use of AI, makers of AI-oriented developer tooling have a shot at a glorious exit. ®

AI has become a juggernaut, driving technical strategy in the tech sector and outside of it. With the worldwide server market growing 30 percent in a single year to hit $122.6bn in the first quarter of 2025, IDC now feels safe recategorizing AI-driven investment "from cyclical to durable".

The analyst company expects inference to account for over 70 percent of computing demand by next year, adding that it's growing fastest at the edge.

All this investment in AI infrastructure might once have come mainly for hyperscalers, but that's changing quickly. Adoption is spreading quickly beyond public cloud service providers into enterprise environments, IDC adds. They're eager to run agentic AI in-house.

There are plenty of reasons for companies to own their own AI infrastructure rather than offloading it to the cloud. One of these is data control, especially in regulated sectors such as finance, healthcare, manufacturing, and the public sector that have sovereign data requirements. Another revolves around the need for predictable costing from tokens delivered on tunable in-house systems.

So what should companies look for in a supplier? Delivery speed will be a powerful driver for many who are under the gun to deliver AI initiatives for the business, especially during periods of constrained memory and CPU supply. Life cycle services will also come near the top of the list, for companies that need help managing everything from the power through to the reasoning LLM models and the workloads that use them. That need for expertise is unlikely to stop; companies will undergo changes in AI strategy as they get to grips with a still-evolving technology.

ASUS expects this AI surge to continue in the datacenter, but elsewhere too, specifically in edge locations (especially industrial use cases) and at developers' desks, as the applications that will power the enterprise get made. It wants to be top of peoples' list for hardware and services at all these places along the AI continuum.

At NVIDIA's GTC show in Berlin on Oct 21-22, it announced a full-stack AI offering that it's calling its Enterprise Agentic AI Solution. It's three-layer affair, featuring:

• ASUS AI Server and Infrastructure: Its selection of AI servers, including several new units.

• ASUS AI Software Stack: Its software for integrating and controlling AI workflows.

• ASUS AI HUB: A services operation that ties it all together.

These bring together some existing assets that ASUS had for enterprise AI users, along with some new ones. The time is right for an end-to-end offering with new hardware because of the way the market is shifting, say ASUS execs.

"ASUS servers help enterprises build high-performance agentic AI infrastructure capable of supporting both complex task execution and large-scale AI agent deployment, accelerating enterprise adoption of agentic AI," says Paul Ju, senior vice president and head of infrastructure solution BG at ASUS.

While generative AI initially spiked demand, the advent of agentic AI has sent compute demand into overdrive. Ju explains that the focus is expanding from the GPU.

"Historically, AI servers have been optimized primarily around GPU compute. But as enterprise AI shifts toward agentic AI, workloads increasingly revolve around a continuous 'agent loop' that is heavily CPU-dependent," he says. That loop takes in task planning, code execution, external tool invocation, and reflection or correction work.

The CPU's increased role in this execution work is creating bottlenecks in conventional AI servers, he adds.

The first such sticking point is single-thread performance, which leaves GPUs idling while they wait for CPUs to complete. The second lies in memory bandwidth. When this is managed on a per-core basis, it can slow things down when many agents run at once.

The multi-die design of many CPUs also penalizes them in AI environments, because accessing data across dies introduces latency, Ju argues. CPUs with cores and cache on different dies suffer latency when data has to cross between them. Maintaining cache coherency exacerbates that problem.

Finally, there's a security risk: a lack of hardware level sandbox isolation on some CPUs make it possible for prompt injection to reach core systems, he warns.

Customers also bring their own challenges as they grapple with different applications for AI. Enterprises evaluating AI server options have to match their server platforms to the projected AI workload profiles, Ju adds. Training a supervised learning model to batch-classify insurance claims takes different technical resources to running a low-latency reasoning employee chatbot.

ASUS wants to help customers solve problems like these wherever they come up. With this in mind, the company had plenty of goodies for onlookers at GTC.

Starting at the datacenter, the company announced both single and full-rack systems built around the NVIDIA HGX Rubin NVL8 eight-GPU server board. It has a 3U hybrid-cooled system, the XA NR11-E12L, along with the 2U 100 percent liquid-cooled XA NR11-E12, both targeting large AI reasoning workloads.

The two Rubin units target different use cases. The 3U hybrid liquid cooling system uses direct-to-chip liquid cooling on the 8GPU baseboard, along with air-cooled dual Xeon 6 chips on a sled. The company pegs this as the easy transition path into AI factory hardware. The 2U unit, with its 100 percent liquid cooling, is tailored for maximum PUE and compute density. It also uses a cable-free board-to-board design to keep its footprint low.

The kit will be speedy, too, with NVIDIA HGX Rubin offering 800Gbit of integrated bandwidth per GPU.

ASUS is promoting the serviceability of these units through their use of tool list CPU and GPU sleds. It claims up to a 50 percent time saving on maintenance operations for these units, meaning less downtime per service and ultimately a lower total cost of ownership.

ASUS is structuring its hardware proposition around speedy delivery so that customers can get hardware up and running to support their AI initiatives. That speed comes from validation carried out before delivery with the help of a pre-validated rack reference architecture.

In keeping with its focus on multiple designs for the right AI workload, ASUS also supports NVIDIA's modular MGX Blueprint for server builds. The selling point here is flexibility with a mixture of GPUs, CPUs, and networking options for enterprise inference, agents, and visual computing. This is the modular route for workloads that don't need the power of an 8-GPU HGX baseboard, and it covers a large gamut of enterprise inference and agentic use cases.

ASUS showcased MGX servers including the 6U XA P8A-E14AXL with its eight liquid-cooled NVIDIA RTX PRO 6000 Blackwell Server Edition GPUs. A 2U sibling, the XA P4N-E2, sports NVIDIA Vera processors for agentic work.

The ASUS booth also featured the ExpertCenter Pro ET900N G3, powered by the GB300 Grace Blackwell Superchip developer desktop with 748GB of coherent memory, and the Ascent GX10 based on NVIDIA DGX Spark for agentic work. Also on show were the ProArt GR1X Mini PC and ProArt P16 and P14 laptop - the first laptops powered by NVIDIA RTX Spark - aimed at both developers and creators.

For edge installations, it showed the PE3000N, its rugged NVIDIA Jetson T5000-based edge-AI computer aimed at use cases including industrial environments. Announced a year ago, this delivers up to 2,070 FP4 TFLOPS of peak AI performance with 128GB of memory. Targeted at industrial AI environments was the PE6000G, a workstation-grade AI and simulation platform.

Equipment only supports an AI factory when there is power to run it. ASUS has been taking steps with NVIDIA as a partner here, adopting its DSX AI Factory platform with digital twin system for laying out hardware and power in AI factory environments. DSX supports equipment using the NVIDIA Vera Rubin processor line, which supports dry coolers and a lower PUE with a 45 degree centigrade warm water inlet.

ASUS configured racks become simulation-ready assets inside DSX, adopting its OpenUSD shared file format. That covers engineering data alongside the geometry for measuring real power draw and cooling requirements.

ASUS wanted to target cooling as part of the original design plan for the AI Factory, not as a facility afterthought. So, the DSX partnership allows implementation teams to model water temperature, flow, pressure, CDU placement, and piping within the digital twin model. They translate information from the planning resources that go around this, such as the NVIDIA Omniverse DSX Blueprint framework, using the ASUS Infrastructure Deployment Center.

The digital twin system isn't just for design and setup. It's supposed to keep working after go-live so that AI factory operators can simulate changes before touching hardware. The ASUS Control Center monitors the infrastructure while DSX exchange links IT and operational technology data. Operators can validate changes in the simulation before flipping the switch to go live.

ASUS' goal here is clear: to assist enterprise customers as they tackle in-house AI opportunities while acknowledging that the performance profile and physical locations of their workloads will vary. As companies continue to evolve their understanding of AI and where it fits within the organization, they will need to be agile in how they plan and nurture their infrastructure from the datacenter out to the edge, and to the developer's desk.

The Berlin announcements tie it together in a neat package for organizations that need help with their agentic aspirations.

Learn More about ASUS NVIDIA HGX Rubin NVL8 Systems: XR R1I12B-M1BLFS | ASUS Servers

Sponsored by ASUS

If you share content using public iCloud links, Apple could soon be looking over your shoulder and checking for "harmful content." The iBiz announced plans to moderate publicly accessible iCloud share links in version 27.3 of iOS, iPadOS, macOS, and visionOS on a support page published on Tuesday. To clarify, this applies only to iCloud content shared via links created to be accessible to anyone, as opposed to content shared directly with other people.

"Items shared using Anyone with the Link will be checked by Apple for harmful content," Apple explained on the new help page without bothering to define what harmful content means. Based on the wording, it appears those checks will apply to all iCloud items shared using "Anyone with the Link," which Apple says it can also access. The company hasn't explained how the checks will work.

The share-with-anyone option removes end-to-end encryption from the shared item, Apple said, even if Advanced Data Protection, Apple's optional end-to-end encryption feature for supported iCloud data, remains enabled for the account.

"If you try to upload harmful content, Apple reserves the right to move, refuse, modify, remove, or delete the content at any time," the help page continues, adding that the iMaker may also block or disable accounts associated with harmful content.

The company also said it reserves the right to involve law enforcement when necessary, suggesting that its undefined "harmful content" category could include child sexual abuse material (CSAM), malware, and other potentially unlawful content.

Without a proper definition of what it defines as harmful, however, iCloud users are essentially left at Apple's mercy: If it chooses to continue to bend the knee to the Trump administration, for example, content deemed objectionable by the standards of political figures could end up wiped out.

Apple also hasn't explained whether human moderators, automated systems, or both will be responsible for identifying and removing so-called harmful content, or whether any decisions will be left to AI, which can easily be fooled into making a bad moderation call.

The scanning of such content, as mentioned above, begins with version 27.3 of Apple's various OSes, which were released to developers in beta on Tuesday, with a general release expected in December.

Apple didn't respond to questions for this story. ®

As of Tuesday, developers can explore expanded Siri AI integration for their Apple platform apps through beta releases of iOS 27.3, iPadOS 27.3, macOS 27.3, and visionOS 27.3. Next spring, nearly three years after announcing plans for deeper third-party Siri integration, Apple expects to begin accepting App Store submissions using the new capabilities.

"This developer beta offers a sneak peek at Siri AI app intents for developers," Apple said in a developer news post. "With this early Developer Preview, your app can work directly with Siri AI, giving you a first look at new ways to reach and serve your customers."

Anything is likely to be an improvement. Apple's Siri, a voice-activated assistant service, dates back to 2011 and gained limited third-party app access in 2016. In 2024, it became an interface for interacting with Apple Intelligence, its suite of generative AI features.

Apple Intelligence has not been very intelligent - several advertised Siri capabilities were delayed, so much so that Apple was sued in 2025 over allegations of misleading marketing. After a shakeup of Apple's machine learning group in December 2025, Siri and Apple Intelligence received a makeover at Apple's Worldwide Developers Conference in June.

"Before this year, Siri was a timer with a voice," said Roberto Gutierrez, Apple platforms staff engineer at a media company and author of the e-book On-Device Intelligence, in an email to The Register. "Weather, alarms, calls. Anything past that was a coin toss.

"On the developer side, SiriKit covered a handful of domains and Shortcuts only worked if users memorized the phrase. Siri AI is the first version that understands context and can chain actions across apps. That's a real change."

There is a catch, however, Gutierrez observed. Siri AI is available in a limited number of languages and regions. Support for Traditional Chinese is expected by the end of 2026, and EU availability remains a matter of negotiation - Siri AI is unavailable on iOS, iPadOS, and watchOS in the bloc as Apple works through interoperability concerns under Europe's Digital Markets Act.

"Apple is working constructively with the EU in an attempt to find a path forward," the company said last month without committing to any concessions.

Gutierrez said that he expects opening Siri to third-party developers will be useful to some extent.

"Siri can only do what apps expose through App Intents, and SiriKit is on its way out," he said. "If your app's main actions aren't exposed, the new Siri can't see you. The bigger shift is letting Claude, Gemini and others plug in. Developers already choose between three options for every AI feature: Apple's system model, an open model through MLX, or a cloud API. Siri is heading the same way. It doesn't need to be the smartest model. It needs to be the best router. That's the right call."

Apple has made choosing between those three options more difficult by failing to publish benchmarks that show how its Foundation Models compare with open-weight alternatives.

The biz has published measurements of how its third-generation Foundation Models compare to their 2025 predecessors. But the company's commitment - to "[share] more details on our latest foundation models, including updated evaluations and benchmarks, in a technical report later this summer" - remains unfulfilled so far as we're aware.

Gutierrez ran his own benchmarks to compare Apple's iOS system model (FM system, ~3B) to several open-weight models that are suitable for use on-device.

"On an A17 Pro iPhone running iOS 26, I tested Apple's system model against Qwen3 1.7B and 4B, Llama 3.2 3B and Gemma 3 4B on a structured-output task," he said. "Speed: Apple's model ran at about 149 tok/s, roughly 3x the fastest open model (Qwen3 1.7B, about 46)."

Quality, he said, was a tie, with every model passing five typed checks. But memory, he said, was where Apple's models had an advantage.

"The system model's weights live outside your app's memory budget," he explained. "My app peaked at 12 MB during inference. An open 4B model costs 2.4-2.7 GB, and a 2.4 GB model plus your app plus a WebView is a crash report. So yes, it's competitive, and on iPhone it's the default I'd recommend. If you want your own model, Qwen3 1.7B is the one."

But Gutierrez urged Apple to publish proper competitive benchmarks. "Their numbers compare against their own previous models," he said. "Developers aren't choosing between Apple 2025 and Apple 2026. We're choosing between Apple, Qwen and a cloud API. I had to build my own harness to answer a question Apple could settle with one table."

The absence of competitive model metrics is not the only gap. Apple has also failed to accommodate web apps.

"Web apps are locked out," said Gutierrez. "Foundation Models is a Swift API (and now Python). No JavaScript, no WebKit. A PWA's only option is to ship its own model into the browser and pay the full memory bill inside a WebView."

And bringing your own model may pose performance problems. "On an A17 Pro iPhone, the 3-4B open models pushed the phone into 'fair' thermal state and dropped to 12-24 tok/s," he said. "In practice, your own model on iPhone means ~2B or smaller."

With Siri AI unavailable on iPhones, iPads, and Apple Watches in the EU (and its rollout in mainland China subject to regulatory requirements), developers may need a fallback for global apps to handle regional restrictions. Apple's Private Cloud Compute offers eligible developers access to Foundation Models, Gutierrez observed, though it requires membership in the App Store Small Business Program, fewer than two million first-time App Store downloads across their apps, and the necessary entitlement. ®

Alex Karp, CEO of US spy-tech company Palantir, used this week's historic US midterm election results to claim an unlikely mantle for a military software supplier: pioneer of an antiwar movement.

Palantir's Q3 revenue rose 96 percent year-on-year to $2.31 billion, although its shares fell 8 percent following the results, which it reported on Tuesday as US voters were delivering a resounding defeat for Republicans in the voting booth. For Karp, the figures represented "the reality that this moment has become a real focal point of understanding" and might help explain "what is happening in our society."

From the CEO's perspective, they not only reflected the company's financial performance, but also its overall "mission."

"That mission reason is we support your alpha [above average performance], we support your sovereignty. We support the war fighter. We are delighted that we can support the war fighter even more because we're making so much money. But we believe we have a moral imperative to be the dominant player in our industry," he told investors.

For Karp-watchers, this is familiar self-serving shtick: turn Palantir's work for military and intelligence customers into a moral crusade, and goad its critics along the way.

"There's this larger question of what does it mean to have a business that was built around what we believed and believe now is the most important mission in the world, supporting our war fighters, [and] bringing them home," he said.

None of this stops him presenting Palantir as a force for peace. "We see ourselves as the pioneers of a genuine and viable antiwar movement," he wrote in his shareholder letter.

He expanded on the point during the Q3 earnings call. "Our products are built to avoid war, meaning the adversary knows that they will have such pain inflicted upon them that, in fact, it's not worth going to war and also to create social cohesion because we're going to bring back our troops healthier, safer, undamaged and that makes our society much more effective," he said.

That antiwar pitch sits awkwardly alongside Karp's $1 million donation in December 2024 to MAGA Inc., the super PAC supporting Donald Trump. Trump subsequently launched a US military operation in Venezuela and went to war with Iran, repeatedly promising the latter conflict would be short-lived as it dragged on. Karp did not address either topic on the call.

Elsewhere, Karp claimed that Palantir "built products that would protect civil liberties and stop terrorism. If you don't stop terrorism, you're going to get a radical fringe on the right. If you stop terrorism without rights, none of us wants that."

That civil liberties pitch also invites scrutiny of Palantir's longstanding relationship with US Immigration and Customs Enforcement (ICE), whose treatment of people it targets and detains has drawn criticism from US judges.

Karp seems to enjoy the circus of controversy he and his colleagues have created. He also offered investors this insight: "The future is becoming the present and the present is now." Whatever that future holds, he seems confident Palantir has a moral imperative to invoice it. ®

Fedora 45 is here, after not one but two week-long delays as the team found last-minute issues. Although both the original and second announcements were irksomely non-specific, in general we prefer this "when it's ready" approach to the rigid calendar-based schedule that Canonical favors for Ubuntu. It includes six main editions, five immutable desktop variants, and 12 "spins" with different GUIs.

The combined version 45 release notes detail some of the bigger individual component changes and note the hardware requirements. The recommended spec is now a 2 GHz quad-core CPU with 8 GB of RAM and 20 GB of disk space, and at least 1024×768 graphics. Fedora 45 supports most display adapters, but 3D acceleration may not work on older GPUs, including Intel models prior to GMA9xx, Nvidia models prior to NV30 (GeForce FX5xxx series), and AMD Radeon models prior to R300 (Radeon 9500).

We took a quick look at the beta back in September, focusing on its new next-generation text console, which is a welcome change. We especially liked the ability to zoom in and out on the text console on the fly, which is something we've missed since the superb "Laptop Ultravision" for DOS, about a third of a century ago. (The company is long gone, but we found a downloadable version for all you DOS users.)

The desktop flagship editions are Workstation 45, which comes with GNOME 51 "A Coruña", and Fedora KDE Plasma Desktop, with KDE Plasma 6.8. We covered some of the new features in GNOME 51 when we looked at the Ubuntu 26.10 beta, and those changes apply here too - notably, improvements in the Settings app, but removal of some older Nvidia GPU support.

Both the big names are Wayland-only environments now, but by our reckoning, X11 fans have seven of the 12 official spins to choose from: Xfce, Cinnamon, MATE, i3, LXQt, LXDE, and Sugar (SOAS) all still use X.org if you prefer. There is also a combined What's New page for all the Atomic Desktops, Fedora's immutable variants - the main shared feature here being the new web-based installer.

The new version of main Fedora is accompanied by Fedora Asahi Remix 45 for Apple Silicon Macs. This is a more notable release than the cross-platform editions, as Asahi 45 brings initial support for M3-based Macs as well as M1 and M2 kit. As The Reg described in September, though, the M3 support is preliminary and limited - for instance, it excludes the M3 Ultra. If you have an M3 Mac, for now, you're better off sticking to macOS.

The most important changes in Fedora 45 are under the hood, which is what the Red Hat release announcement focuses on. It now uses RPM version 6.1, with enhanced package signature checking, and blocks the vendor of packages silently changing. The GNOME keyring now uses the oo7 secrets manager. Support for two old VPN plugins has been removed, because the packages were unmaintained.

A feature that we are happy to see land at last is better support for Red Hat's Stratis next-generation storage subsystem. We gave a potted overview of Stratis 3.3 back in 2022: it was on version 3.3 back then. It's the same main Stratis version 3.9 we saw in Fedora 44, but the installation program now has support for installing Fedora 45 on Stratis storage.

Stratis is Red Hat's rival to OpenZFS. The plan is that Stratis offers similar features and resilience to ZFS, especially in the field of copy-on-write snapshots and simple multi-device logical volume management, but built from all-GPL code. Rather than a whole new filesystem, Stratis integrates XFS, LVM, and the kernel device-mapper functionality. Stratis 1.0 appeared in Fedora 29 eight years ago, and Fedora 45 has version 3.9.3. It's good to see it finally making real progress: if nothing else, we've seen Btrfs irretrievably corrupt a full volume far too many times to ever trust it again, so an alternative is very welcome.

When we looked at Fedora 44, we noted the difficulty of doing custom disk partitioning. It was there, but very well-hidden. Now, the installer can launch the Cockpit web-management tool to make it easier to set up your drives however you want.

An interesting side-effect of the delayed release is that Fedora 45 ships with kernel 7.2 - specifically, version 7.2.15, which is the last release of this short-term series. In other words, it's an end-of-life kernel version. It's an amusing contrast to its near-contemporary Ubuntu 26.10, which shipped with a pre-release kernel. Our test Ubuntu box was automatically upgraded to the release version by the background automatic updates process, but to our surprise, it didn't receive the kernel update for over a week. This shouldn't be a problem: unlike Ubuntu, Fedora releases routinely receive updates to whole new kernel versions, so we're sure Fedora 45 users will get kernel 7.3 very soon.

We upgraded the same test machine on which we tried GNOME 48 in Fedora 42 early last year. You can skip one Fedora release on upgrade, but not two, so we had to go via Fedora 44 — which took some time, and threw up a minor packaging glitch, but it got there. The result runs surprisingly well, although this 2008 machine is well below the recommended spec. ®

Your new robot servant doesn't have to look anything like a person or walk on two legs. In fact, it might just be a claw that hangs from the ceiling instead.

That's the future envisioned by Nathaniel Nifong, founder of Neufangled Robotics, and its Stringman robot. Put simply, Stringman is an AI-equipped claw suspended from four strings running through the corners of a room (up to 10 meters diagonally). Two motorized anchors adjust the lengths of the strings to move the claw around the room, from ceilings up to 4 meters high down to the floor to pick stuff up. With a payload capacity of around 2 kg, depending on its position, Stringman can move items to designated locations - like a toy box - and perform tidying tasks.

Stringman can also be manually controlled if you'd prefer to play a room-sized crane game by yourself.

According to Stringman's documentation site, the device is currently a public prototype that "can self-calibrate in a new room and run policies, but is not yet reliable enough to be called a home appliance."

That said, demonstration videos on the website show it in action, and it's neat to watch.

Stringman is shown picking up laundry and dropping it in a basket, putting Lego bricks and other kids' toys back where they belong, and generally zooming around rooms and demonstrating its agility. One could imagine a properly trained Stringman could even grab the remote you left on the other side of the living room and drop it in your waiting hand if you're that incredibly lazy.

"Our mission is to give people an organized home," Neufangled says on its future roadmap page. "We prioritize robots that are narrowly effective at something that helps today rather than a general purpose robot that may be in development for years."

The company further describes its goal as designing robots that are low cost and capable of doing things that people hate to do - e.g., pick up after a messy toddler - and making a robot that can do that while camouflaging itself into the home "as well as possible."

As long as you think a foot-long claw robot hanging from the corner of the ceiling is camouflaged, this is the robot for you.

Neufangled lists Stringman for $795, which is a lot to ask for a public prototype that isn't reliable enough to be called a home appliance yet. Luckily for those who're interested in this sort of niche home automation, it's fully open source, too.

"A complete Stringman is three Raspberry Pis on your Wi-Fi network," the documentation site reads: One in the gripper, and two in the corner anchor mounts that store line reels and have cameras to track the claw's location as it moves around the room. A third camera mounted on the claw provides visual feedback. Oh, you'll also need a dedicated PC or Mac to run the robot's motion-control software.

As for Stringman itself, to build your own, you'll need a 3D printer, a supplier to manufacture the PCBs, motors, Raspberry Pis, and other components listed in the bill of materials. You'll also need to follow the guides on the documentation site to get it up and running, calibrate it for your space using printed markers, set up drop-off points using visual tags, and program it to perform tasks.

But is it safe? One could imagine getting tangled in the lines if you get in Stringman's way. Nifong admits that can happen, writing in the documentation that he has walked into the lines himself without injury.

"I live with two of these, and I've walked into the lines before. It's startling, but harmless — they're light and they give," he writes. According to the robot's website, it has a tension limit of 28 newtons, or around six and a quarter pounds of force. If that limit is reached, all of the lines relax and can be pulled loose.

Those who want to play with a simulated Stringman can do so on Stringman's web portal, which includes options to connect to a personal machine over LAN, the cloud, or a simulated version. Hopefully it's easier to work with than an arcade claw. ®

Scientists have developed a way to remove starlight from telescope images that improved sensitivity to faint exoplanets by up to 15 times in tests on James Webb Space Telescope data.

A team led by Northwestern University's Rodrigo Ferrer-Chavez combined a physics-based telescope model with optimization techniques used in machine learning to subtract unwanted starlight.

Directly imaging exoplanets remains difficult because their host stars' glare can overwhelm their faint light. Existing techniques subtract modeled or reference starlight, but tiny changes in telescope optics can leave residual glare that hides planets.

The team's Physics-based Learning for Imaging Planets (FLIP) framework models how light passes through the telescope, adjusts for optical imperfections, and simulates the glare to be subtracted.

Applied to observations from Webb's Near-Infrared Camera (NIRCam), FLIP produced the first image of the known giant planet 51 Eridani b at an infrared wavelength of 4.3 micrometers, where conventional processing had failed. Tests showed sensitivity improvements of two to 15 times over conventional techniques, according to the Nature paper.

"This is the first time, to our knowledge, that such an approach has demonstrated substantially improved sensitivity on real, on-sky coronagraphic imaging. Our framework can be extended to other current and future ground-based and space-based observatories… potentially relaxing the unprecedented stability requirements needed to make these a success," the paper says.

Further work includes reducing computational demands and developing ways to handle the much larger data volumes produced by ground-based observations.

The researchers hope further development and testing will bring direct images of Earth-like planets closer, while reducing the extent to which tiny telescope instabilities obscure them. ®

On-premises high-performance compute (HPC) is still in demand, according to Hyperion Research, which estimates the market is expanding at almost 20 percent year-on-year, often driven by requirements for AI processing.

What Hyperion calls the "on-premises HPC and AI-for-Science" server market showed significant growth in the first half of 2026, with revenues hitting $16.5 billion, up 19.9 percent.

However, the analyst says unit shipments over the same period grew only 4.4 percent, meaning that systems are carrying a higher price tag due to greater use of costly components such as GPUs and other accelerators, not to mention the effects of higher memory costs, which have tripled or quadrupled since last year.

Its tracking of the HPC market is based on multiple surveys conducted of both buyers and suppliers, the firm says. The latest data shows a major change in the overall HPC market growth, heavily driven by a combination of the use of AI, large language models (LLMs), and purchases of large GPU/accelerator-based systems.

This trend was noted a couple of years ago, with Omdia reporting in 2024 that revenue growth from server sales was racing ahead of actual units shipped, as customers prioritized investment in highly configured server clusters for AI.

Much of that spending was by hyperscalers and cloud providers, seeking to profit from selling AI services to enterprise customers, but Hyperion's figures focus on on-prem shipments, showing that many enterprises prefer to do their processing in-house.

According to Hyperion, Dell has now surpassed HPE as the biggest HPC vendor, with server revenues of $4.2 billion for the first half of 2026. HPE, which picked up supercomputing specialist Cray for $1.3 billion in 2019, had long been pack leader in the global HPC arena in terms of market share and revenue.

The figures put Dell on a 24 percent share, while HPE is not far behind on 21 percent. Lenovo and China-based Sugon trail behind at 4 percent share each, while 45 percent of the market is just labelled as "others" by Hyperion. This includes names such as Bull, which re-emerged from the shadows of Atos this year, plus Supermicro, Penguin Computing, and NEC.

The HPC infrastructure market is inherently cyclical. HPE recognized significant revenue in 2025 with sales of multiple exascale-class leadership machines, but procurement of that scale is intermittent and doesn't necessarily come along every year, Hyperion notes.

"Organizations and institutions are continuing to invest in their advanced computing infrastructure due to both replacing aging machines and the desire to adopt AI into their workflows to support their businesses and missions," said Hyperion Research CEO Earl Joseph.

"The growth reflects both a nominal increase in units being purchased and significant increases in system prices."

HPC systems are considered the backbone for many AI and AI-related applications, which has driven the overall HPC market to a new growth level, Joseph added. ®

As first reported by The Register, VMware has returned to the low-end server virtualization market with a product called "vSphere Express."

The product will run on up to 128 processor cores and includes vMotion to allow live migration of VMs, automated restart of failed VMs, a distributed switch, rapid non-disruptive patching, and support for physical and virtual trusted platform modules.

"We have been listening to the market at the low end, and they want a basic core server virtualization platform," said Prashanth Shenoy, vice president of product marketing in Broadcom's VMware Cloud Foundation (VCF). He explained that each instance of vSphere Express will be discrete, meaning it won't be possible to manage them centrally.

"This is tested for small implementations," he told The Register. "We do not want a customer to run this on thousands of cores." Shenoy said the vSphere Foundation (VVF) enterprise virtualization product, and Cloud Foundation (VCF) private cloud suite, are appropriate for users with more complex needs. vSphere Express will, however, match the release cadence of VVF and VCF.

Under Broadcom's ownership, VMware has enormously reduced the number of resellers in its channel program, meaning it no longer has partners who are a good fit for this product.

The virtualization giant will therefore sell vSphere Express through its distributors, most of which now operate cloud marketplaces at which affiliated resellers can acquire software - an arrangement channel folk call "two tier distribution."

We asked VMware to tell us the price it will charge for vSphere Express.

A spokesperson told us it will offer a per-core subscription license at a price "significantly lower" than VVF or VCF but declined to quote a dollar amount. "We encourage interested customers to reach out to their local reseller for information related to pricing and subscription terms," the spokesperson wrote.

VMware has also bulked out VVF and VCF a little, by adding disaster recovery and policy-based networking for virtual private clouds that it previously sold as add-ons. Shenoy pitched the additions as demonstrating VMware's intention to add value to VVF and VCF, in part by ensuring the bundles include tools that users are likely to consider represent essential functionality.

The Broadcom business unit will also add new features, such as native AWS S3-compatible object storage for its VSAN virtual storage. That update, due in calendar Q1 of 2027, means VSAN will become a more convenient platform on which to store the unstructured data that AI workloads rely on, therefore boosting the utility of VMware's AI Factory.

VMware also plans to make its AI Factory easier to implement.

"The idea is to focus more on AI, less on setup," Shenoy said.

VMware's return to the low-end server virtualization market may not be well-received, because it has largely ignored users of its existing low-end vSphere Standard product for the last few years, and irritated many by insisting they upgrade to the more expensive and complex VCF.

The virtualization giant also cast aside thousands of resellers.

When The Register chats with members of the VMware community, we often hear that trust in the vendor has diminished.

Your correspondent therefore feels that vSphere Express will get a mixed reception.

Many vSphere Standard users will welcome a more appropriate and affordable upgrade option, especially as it means they won't need to learn new skills.

Other users and former partners will wonder if, having been bitten by Broadcom once, they should now be a little shy of vSphere Express.

Broadcom shareholders should be enthusiastic, as VMware's decision not to sell to the low end of the market blocked its onramp to new users. The chips 'n' code shop now has another way to get VMware growing, if it can restore trust. ®

He swore an allegiance to the US, but he was working for someone else. An IT specialist who worked across US government departments and held Top Secret clearance has pleaded guilty to transmitting sensitive documents to Ethiopia.

Abraham Teklu Lemma, 53, admitted to the espionage offenses and agreed to a prison term between 120 and 262 months, the Justice Department confirmed on Tuesday.

A naturalized citizen of the US, Lemma, of Maryland, had worked for the US government in different capacities and across different teams, including for the National Reconnaissance Office, State Department (DoS), and Justice Department (DoJ), between 2019 and 2023.

These roles included working evening hours as an IT help desk technician at the DoS as of 2021, and daytime work at the DoJ as a contract management analyst, according to court documents.

Lemma had what's known as "downdraft" authority, which is to say that he was permitted to move classified files between classified systems, and unclassified files from classified to unclassified systems.

Crucially, he was only allowed to access classified files on non-DoS systems with written approval that it was part of his official duties, and worked in a Sensitive Compartmented Information Facility (SCIF) - a secure workplace built for handling classified files.

The criminal complaint [PDF] initially stated that the alleged offending began in July 2022, although the DoJ more recently claimed that it began in 2021.

It claims Lemma was caught trawling "several" non-DoS classified portals and accessed intelligence reports without the proper authorization to do so.

On various occasions, Lemma, who had held Top Secret security clearance since 2020, copied information from at least 133 intelligence reports without need-to-know authorization, according to court documents.

Prosecutors claimed that the man downloaded more than 100 documents, burned them to CD/DVDs, and deleted their classification markings.

The material Lemma was copying was at the Secret and Top Secret levels, but burned onto discs with green labels, which are approved for employees and are supposed to signify the contents of the discs relating to unclassified material only.

Lemma would allegedly take these discs at the end of his shift and exit the SCIF with them tucked into his jacket pocket.

On other occasions, he was observed looking at classified material on his PC, then taking handwritten notes before folding the paper and again storing it in his pocket. This continued until around August 2023.

Prosecutors claimed that Lemma began transmitting information to Ethiopian officials via an encrypted messaging app in or around August and September 2022, weeks after the alleged offending began.

The files included documents, photographs, notes, maps, and information relating to Ethiopia, and officials would ask Lemma to provide details about command centers, logistics centers, military compounds, and other matters.

Communications also covered geopolitical topics, referencing "situations" in an unspecified African country that shares a border with Ethiopia to gain an advantage, and rebel groups involved in conflict with Ethiopia.

Lemma's messages also included encouragement to take action against paramilitary groups that were in conflict with Ethiopia on multiple occasions, say prosecutors.

According to the complaint, one of Lemma's messages said: "We have to finish these people once and for all, otherwise there will never be peace in Ethiopia." Later that day, on December 18, 2021, he wrote: "Please send dro[nes]," after sharing additional information.

Announcing the guilty plea on Tuesday, the Justice Department said many of Lemma's contacts were senior Ethiopian officials across political, military, and intelligence branches of the government.

Lemma was also accused of storing copies of classified files on his own storage drives, which prosecutors claim were accessed remotely by Ethiopian officials numerous times between April and June 2023.

The man was additionally caught making several bank deposits exceeding $66,000 throughout the time he was working on behalf of foreign officials.

On one occasion, he was observed "angrily discouraging" bank staff from filing a currency transaction report after trying to deposit a sum greater than $10,000.

"Despite swearing an oath to defend the Constitution of the United States, Abraham Lemma betrayed our country by committing espionage on behalf of the Ethiopian government," said Roman Rozhavsky, assistant director at the FBI's Counterintelligence and Espionage Division.

"By his own admission, Lemma provided classified national defense information to Ethiopian officials, who then used it in military operations during an ongoing conflict. This guilty plea should serve notice that the FBI and our partners will work tirelessly to stop foreign intelligence services, and anyone who decides to work with them, against the interests of the United States." ®

HP Inc has hired former GSK boss Emma Walmsley to plot its course through choppy PC and print waters.

Walmsley spent eight years as the boss at pharma biz GSK and further back in time held various marketing and general management positions at L'Oreal. She has no track record running a tech company, though she was on Microsoft's board.

"Throughout the process, Emma stood out for the strength of her leadership experience across multiple industries, the clarity of her strategic thinking and her ability to translate customer insights into new growth," said Chip Bergh, chair of HP's board.

In the first nine months of HP's fiscal 2026, revenue had jumped 10 percent year-on-year to $44.5 billion. The print business shrank in that time and PCs generated the growth. However, amid a memory crunch with supplies dwindling, analysts are forecasting PC market contraction in terms of unit sales.

Global PC shipments slumped 20 percent in Q3 as DRAM shortages forced up prices and customers opted instead to sweat their assets rather than replace their fleet of computers. Things aren't expected to improve in the final three months of 2026.

One beacon of hope for all PC makers was AI, yet many customers aren't buying into the story for various reasons.

Walmsley previously said that during her time at the top of GSK she created a "totally different company, in platform, in momentum, in talent, in culture," she told the FT. "I think in time… it will be recognized." She also dealt with activist investor Elliot Management, which operates in the tech industry and called for her to reapply for the CEO role at GSK, claiming her lack of science background was an issue.

Walmsley said: "HP is an iconic company with a defining role in shaping the future of work on a global scale. I'm excited to join the team and, together, move forward with clarity and conviction. We will focus on harnessing the power of the AI era and putting HP's creativity and innovation to work to unlock new sources of growth and deliver lasting value for our customers, shareholders, partners and people."

Employees will be crossing their fingers that their new boss doesn't borrow from the playbook of her predecessor - who hotfooted it to PayPal in February - by running multiple redundancy rounds, and shareholders will hope their stock price show sustainable improvement. If not, Walmsley may have to manage more interest from activist investors at the latest corporation she is running. ®

Datacenter restrictions found broad support among candidates for governor in this week's US midterm elections, with nearly all those examined by Data Center Watch backing bans, pauses, or tighter development rules.

Data Center Watch, which tracks grassroots opposition to compute sheds across America, examined the public positions of 74 selected candidates across the 36 states that elected governors in the midterm elections on November 3.

It found that 43 percent of those candidates supported a ban or moratorium on new projects, while another 51 percent backed further regulation or conditional development. Only about 5 percent had no stated public position.

Support for bans or moratoriums crossed party lines, although it was more common among the Democratic candidates examined (60 percent) than Republicans (30 percent).

Candidates from both major parties backed restrictions in eight states: Maryland, Maine, Michigan, New Hampshire, Ohio, Oregon, Wisconsin, and Texas.

Texas has experienced a boom in data campus development recently, thanks to its ample energy resources and light-touch regulations, although outgoing governor Greg Abbott ordered a pause earlier this year while audits of energy and water use were carried out. Democrat Gina Hinojosa defeated Abbott in Tuesday's election and will take office in January.

Those positions matter to developers because governors can shape state regulation, energy policy, and permitting, although approval powers vary by jurisdiction.

However, temporary pauses and targeted construction restrictions do not necessarily amount to opposition to all datacenter development.

AI-related campaign spending has also come under scrutiny. Guardrails Alliance, a labor-backed pressure group, described Leading the Future, a super PAC network opposing AI regulation, as "largely a non-factor" in the elections. It alleged the network avoided the most competitive races because "they know their money is politically toxic."

Anti-datacenter sentiment has grown in the US in response to the facilities popping up in people's backyards, often without them having any say in the process.

A Gallup survey earlier this year discovered that more than 70 percent of Americans would be against the construction of one in their neighborhood, while a Pew Research Center survey found that Americans increasingly believe datacenters are bad for the environment and blame them for rising household energy bills.

Pew also found that advertisers spent almost $12 million on Facebook and Instagram political ads mentioning datacenters between July 27 and October 24. Among those running in October through the study's cutoff, 57 percent opposed datacenters and just three percent supported them.

Common themes included corporate interests, energy use, environmental effects, and water consumption. ®

Capita has repeatedly "broken" promises to resolve the "chaos" engulfing the Civil Service Pension Scheme (CSPS) following its takeover of the administration - a move that left thousands of retired public sector workers unpaid.

So says Parliamentary spending watchdog, the Public Accounts Committee, which this week launched an inquiry into the handover of the ten-year £239 million contract in December from previous incumbent MyCSP.

As The Register exclusively revealed, the troubles began early on as members of CSPS - which manages pensions for 1.7 million people - complained of unrecognized passwords and usernames on the portal Capita developed. Capita said it inherited 132,100 complex remediation cases, and claimed the work-in-progress backlog from the prior system was more than double the 37,000 items when it won the deal in 2023.

Capita ran to Microsoft for support, and promised the use AI and automation, in addition to its 500 full-time employees working on the project, would resolve the issues by March. It missed this deadline, as well as one set for June 30 and is set to miss March 2027.

The government subsequently withheld a £10 million payment to Capita over the CSPS fiasco.

The National Audit Office, a spending regulator, has already launched its own probe into the failures and is now joined by the PAC, which says thousands of civil servants and their families have struggled to on and collect their pensions

The PAC has already held three evidence sessions on Capita and the government's efforts to ameliorate the situation. The law was changed in the Budget 2026 to mitigate the impact of administrative delays in CSPS on bereaved families.

Sir Geoffrey Clifton-Brown, chair at the PAC, said in a statement:

"In response to the alarming reports of hard-working civil servants being thrown into financial uncertainty following Capita's takeover of the CSPS, we have held a number of evidence sessions with both the top leadership of Capita, and the civil servants and ministers of the Cabinet Office who have ultimate responsibility for overseeing the contract.

"Over and over again, promises have been made and broken by Capita, to our faces in formal evidence sessions, that they will get a grip on what has gone wrong. As a result, we are launching an inquiry, which will be informed by the authoritative factual basis provided by the NAO, which supports the scrutiny work of this Committee."

The inquiry will draw on the finding of the NAO's forthcoming report on the switch of CSPS to Capita, and will take into consideration the status of the scheme at go-live date, how Capita has performed, and the effect on CSPS members, plus take a look into the way the Cabinet Office has overseen the prior scheme and the switch to Capita.

The Register has asked Capita to comment. ®

Attackers tampered with cold-storage settings used by France's civilian blood service after compromising a refrigeration supplier, the organization said.

The French Blood Establishment (EFS) confirmed the incident on Tuesday, saying its own systems had not been directly targeted. Safeguards prevented any loss of blood products.

The affected equipment stored blood samples and frozen plasma, including plasma intended for transfusion.

"The EFS' monitoring and alarm systems enabled the rapid identification of these anomalies and the implementation of necessary measures to ensure that no blood samples or products were compromised," it said.

"Thus, depending on the circumstances, as soon as the incident occurred, the affected equipment was either brought under control without impacting the products and samples, or neutralized; the products stored in these cold rooms were preventively transferred to other facilities."

The EFS did not name the supplier publicly.

According to reports, attackers changed cold-room temperature settings from -30°C to 110°C. That does not mean the equipment or its contents reached the higher temperature.

The organization added that it notified the country's cybersecurity agency, ANSSI, and the Ministry of Health.

News about cyberattacks targeting refrigeration devices across Europe emerged on Monday morning after retailers arrived to work, only to find puddles on the floor after cyber criminals activated defrost modes.

Reports of the supermarket attacks describe remotely activated defrost cycles and temperatures as high as 50°C, forcing retailers to discard spoiled food.

The attackers have not been identified. More retailers have reported disruption since the initial disclosures on Monday.

Stores across Belgium, Ireland, Luxembourg, and Sweden were the first to report issues, although cases from Estonia, Finland, Germany, Portugal, Romania, and the Netherlands have since been confirmed.

Richard Horne, chief executive of the UK's National Cyber Security Centre (NCSC), told BBC Radio 4's Today program on Wednesday that he was aware of "a couple of impacts" in the UK's own food supply chain.

No affected UK businesses have publicly identified themselves, although the British Association of Independent Meat Suppliers (AIMS) told the BBC that some of its more than 300 members had become "collateral" in the attacks, with two reporting that their fridges had begun defrosting.

The latest wave of cold-store attacks comes a month after the first, which affected French supermarket chains Carrefour, Coopérative U, Intermarché, Monoprix, and others.

Both supermarket waves involved remote interference with refrigeration controls, causing stock losses. Whether the attack affecting EFS used the same access route remains unclear.

Some affected chains identified Danfoss controllers in their systems; others said they did not use the company's equipment.

CERT-FR published an advisory at the time of the attacks urging organizations to remediate vulnerabilities in Danfoss AK-SM and AK-SC controllers.

The attacks in France followed reports of compromises at US military commissaries, which also led to food wastage in August. ®

Fresh on the desktop this week is a new version of Enlightenment as well the continuation of noughties hero KDE 3, the Trinity Desktop Environment, for those who prefer things a little old school.

Enlightenment version 0.28 is the latest release of one of the older Linux graphical environments, which first appeared in 1997. Version 0.28 arrives nearly two years after the previous release - we reported on Enlightement 0.27 in January 2025. If you're not familiar with this venerable window manager, often referred to as just E, we refer you back to that article, which also explains Enlightenment's slightly strange version numbering scheme.

Last time around we noted that its single-line release notes really didn't give us much to write about. The development team has made up for it this time, with some 68 lines of changes. It's both the same old E, and simultaneously a modern tool: it is still an X11 window manager, but it can also act as a Wayland compositor if you prefer. It offers conventional stacking resizable windows, but also can act as a tiling window manager if you prefer. On recent distros, it can now talk to the Pipewire audio server, and it can also work with NetworkManager to control network connections. This version delivers preliminary support for NetBSD too.

This isn't really a tool for newbies or those looking for a simple easy desktop. A sign of its age is that the first-run wizard asks if you want windows to activate as soon as you move the mouse pointer over them. This was traditional 20th century Unix behavior, and in the early days of Linux it was called "focus-follows-mouse," but it's rarely seen today. E isn't a full desktop environment: it only offers a few of its own optional tools, such as the Ecrire text editor, Ephoto picture viewer, Evisum task manager, and Rage video player. Even its taskbar is optional. It is more like a basis for experienced users to customize and build their own environment around.

The Trinity Desktop Environment has a rather lower learning curve: TDE is a port of the final version of KDE 3 to modern Linux. Proving old school is still cool, the project - which keeps the look and feel of 2005-era KDE 3.5 alive - has just announced TDE release 14.1.7, with dozens of bug fixes and plenty of new features. Among these is a tweak to the kicker/taskbar: you can now scroll the taskbar using the mouse wheel, flicking through your open applications.This release drops support for Ubuntu 24.10 "Oracular Oriole" but adds support for the latest 26.10 "Stonking Stingray."

Forks of older versions by users disenchanted with new versions are a recurring feature of the FOSS world, but they often tend to peter out after a while. TDE is one of the exceptions: we first reported on a new release in 2021 and again in 2023. ®

Microsoft wants enterprise customers to start thinking about PC refreshes using the same mindset - and line in the budget - as AI investments. The problem? Rising component costs caused by the proliferation of AI datacenters mean many clients are delaying client refreshes.

In a call to arms, Microsoft became the headline sponsor of the Canalys Forum in Barcelona early last month for the first time ever. Redmond needs distributors and resellers that convene at the event annually to get behind Copilot, Cowork and, more generally, AI computers.

Mark Linton, corporate veep for Windows and Device Sales, took to the stage to preach to a room he is hoping to convert.

"What we're seeing is the tokenomics of [AI] is getting very challenging," he said. "Having agents working on frontier models quickly becomes very expensive."

Redmond's proposed solution is to use those massive LLMs where most needed, "the right model for the right job," running smaller models locally at the edge on Windows PCs. "My goal, if we're successful in pivoting Windows to the agentic platform, is to have PC spend be part of the AI spend versus just PC refresh budget," he said.

"In an AI hybrid world, you'll be able to use edge devices, powerful PCs with lots of memory, lots of capability to run your AI workloads that are appropriate right there on the device. You keep the data locally. Of course, you don't have the token cost of the cloud. You have better latency, and you have more control."

Copilot is merely a component. This is not only about customers using Microsoft "first-party software" - it's for all models and all agents to use Windows as the platform, added Linton.

Goodness knows, Windows does need a shot in the arm. Microsoft last week reported profit and loss accounts for Q1 of its fiscal 2027, ended September 30, and revenue in the Windows OEM and Devices division shrank 22 percent from last year to $3.55 billion. Microsoft warned revenue generated in its Q2 may be down 30 percent.

"Growth will continue to be impacted by elevated inventory levels, PC market demand and a prior year comparable that benefited from Windows 10 end of support," Microsoft CFO Amy Hood said on an earnings call.

Market watcher IDC said calendar Q3 PC shipments sank to 62.7 million units, down 20.1 percent year-on-year, translating into 15.8 million fewer sales into distributors' warehouses. Tough macroeconomic conditions coupled with rising memory costs mean tough times for the PC industry.

Omdia - which owns Canalys - said: "The impact of upstream component cost increases has now fully worked its way through to end-market pricing, with consumers and IT decision-makers responding accordingly."

Reg readers know why component costs are going up. The datacenter construction frenzy meant memory chip makers could make more money from higher-margin HBM than conventional DRAM, prompting them to prioritize production of the former.

But AI PCs have been a tough sell from the start, as higher prices and a lack of killer applications meant many customers delayed purchases. Microsoft has to convince the market of the benefits - including trade customers as well as those in charge of procurement at enterprises.

Back at the Canalys Forum, Microsoft's Linton claimed customers mulling AI PC purchases want better productivity, to reduce costs, to innovate, and to be able to use these computers securely and at scale.

Microsoft is going to invest heavily in Windows from both the platform capability and security to enable these ends. So, if you are a Microsoft partner, you'll be able to run Windows and secure agents at the operating system level with Agent 365, with Intune, or with Entra.

Linton added: "We announced in June some of the technology in Windows, in particular Microsoft Execution Containers (MXC). And so essentially you can virtualize agents, and then you can use Intune to give them policy. And so other platforms, you have to manage agent by agent. Windows, you'll be able to do it policy-based, much like you manage people and network resources.

"You will have agents for all, and you need to know how to manage them, track them, audit them, and then ensure that they're contained in the right way. So we're doing that at the operating system primitive level."

Some of the quality control issues that have dogged Windows may give some sysadmins pause for thought about the grand claims made by the Windows exec.

Still, Microsoft is on a mission, and it knows how to attract thousands of resellers by dangling financial incentives in their direction. Linton said it is pumping money into channel programs, including AI-readiness funded headcount with "top partners" to "get this hybrid AI story moving."

Microsoft wants the channel to deploy AI internally too, "especially for your sales team" because "if they're using it, they're able to sell it."

Dell isn't hopeful of an uplift in PC demand this year. David Kennedy, chief financial officer, said sales will remain "compressed."

"The reality is most enterprises this year for their IT budgets are sweating those assets out a little bit more and what dollars they have, they're trying to build out their infrastructure side," he told the Citi TMT conference in September.

"As we go forward, I think you'll see it [become] a little bit more robust. Our main focus will be on those higher ASP (average sales price) commercial boxes as part of the AI storyboard that's there."

He reckons there are more than 400 million PCs owned by customers that are "over four years old", and with the "cheapest token being closest to the data, none of those installed base PCs will have the battery life or the power to do AI workloads on them. So there is a pent-up demand that we believe will come once it stabilizes and once IT budgets get a little bit more settled and expanded."

Will 2027 be the year of the AI desktop? The Reg suspects not, though readers can do what they always do and speak their brains in the comments section. ®

Schneider Electric sees no slowdown in its datacenter business, but says operators must be more open with communities about the power, water, and land their projects will consume.

As a supplier of electrical infrastructure, cooling, and power distribution equipment, Schneider has a direct view of operators' spending plans - and a commercial interest in continued construction.

Steven Carlini, vice president of innovation and datacenter, said Schneider's pipeline showed no slowdown. "And Schneider has a lot of agreements, supply chain agreements with the larger datacenter operators."

Reports earlier this year suggested only half the capacity planned for 2026 is actually under construction in the US, and land being cleared for new bit barn projects has plateaued.

Carlini nevertheless said US companies were building as fast as they could, contrasting that with planning bureaucracy and other obstacles in Europe.

"But the one thing that I think is going to be the biggest change is the level of transparency from the operators when they're trying to secure permits and when they're requesting power," Carlini said.

"The way the model traditionally worked in the US is the operators would negotiate with the local jurisdictions that were responsible for issuing the permits, and there would be non-disclosures involved, almost universally. So the residents where these datacenters were going to be built had no visibility into these projects until construction started."

Opposition to datacenter construction has grown in the US. Campaign groups called for a moratorium on new facilities late last year, and New York state imposed a halt this summer.

The sheer scale of the new builds has also alarmed many folk, with Meta's Hyperion facility, for example, planned to reach 5 GW of power capacity and cover an area equal to a significant chunk of Manhattan.

A recent survey by think tank Pew Research Center found that Americans are increasingly wary of datacenters, with more than half saying they are bad for the environment, bad for household energy bills, and consume water resources for cooling.

In response, companies building and operating compute sheds need to be more upfront about what effect their plans will have on the local area, Carlini said.

"They're going to have to disclose where the power is going to come from, how they're going to get the power, their water usage, their carbon emissions, the efficiency of the datacenter, where it is going to be located physically, the noise levels of the facility, the water sources, which is a big one, and how much water they're going to use is another big one that operators have traditionally not been very forthright in publishing."

In Virginia - the world's "datacenter capital" - an ex-water sustainability program manager at AWS is taking his former employer to court over claims it published false and misleading statements about water use in the state.

Carlini expects permitting requirements to force greater disclosure, followed by reporting obligations once facilities are operating.

"And then they're going to have to report it as part of regulations during operations as well, so it's not going to be a case of here's our prediction, and then you get the permit and you never have to report again," he said.

Datacenter owners and operators must win hearts and minds as well, Carlini added, a point already recognized by some in the industry at an event last year.

"From a public relations perspective, that's another area where the operators really need to focus, because in many cases they've taken the position that they're benefiting the community," he said. "Those days are gone, so the operators are going to have to, from a PR perspective, engage with the local residents more amicably."

People using Facebook or Gmail doesn't mean they'll welcome a datacenter next door, Carlini told us.

"So it's going to have to be more like 'Here's how we're going to help,'" he said. Operators would need to acknowledge residents' concerns and explain any benefits, such as tax breaks or lower costs.

For AI facilities, Carlini expects more attention to token-processing performance and efficiency, rather than power capacity alone.

"I think we're going to start to see more and more focus on tokens per watt, tokens per dollar, and even datacenters could be sized in tokens as a performance metric for the IT rather than just the power metric," he told us.

"I think the industry has suffered in the past because datacenters have always just been sized in power consumption, and that gave little idea what it was producing and what value [it] was adding."

For more on the economics of tokens in AI processing, readers can refer to our article on this topic from earlier this year. ®

China's crewed space program has outlined plans for a moon landing it hopes to achieve by the year 2030.

Deputy chief designer Wang Xiang explained the plans last weekend at an event called the Seventh Manned Space Symposium. Chinese state media has posted video of Wang discussing plans to reach Luna by launching two Long March 10 rockets.

One will carry a lander named "Lanyue" and the other will hoist a crew capsule called "Mengzhou." The two craft will meet in orbit around the Moon, where two of the three Taikonauts aboard Mengzhou will transfer to Lanyue and descend to set boot on regolith.

In the video, Wang said China plans near-Earth orbit test missions for both vehicles in coming years, followed by an uncrewed visit to the Moon. If all goes well, China hopes to attempt a crewed landing by 2030.

The Seventh Manned Space Symposium marked the 70th anniversary of the founding of China's space program and was also pitched as a moment for articulation of the nation's ambitions in space under its 15th five-year-plan, Beijing's blueprint for the years 2026 to 2030.

China called off its last moonshot in August, when authorities decided the Chang'e 7 robotic ice hunting mission "failed to meet launch criteria" just one day before its planned launch. The orbit required to reach the South Lunar Pole for that mission isn't always easily accessible, so China plans to try again next year.

When China tries to get Chang'e 7, Lanyue, and Mengzhou flying, it will at least have some new standards to apply to their operations after the China Manned Space Engineering Office last week published seventeen of them. One of the standards defines the architecture for Ethernet and Wi-Fi networks on China's space station and other orbiters. Another governs signage used in orbit, in the hope that a standard means Taikonauts won't make mistakes. A couple outline procedures for extravehicular activities, and use of a robotic arm to assist Taikonauts on such missions. Others explain the materials designers can choose from when building spacecraft. ®

AMD has predicted it will win more revenue than Intel in its next quarter, potentially marking the first time it becomes the dominant vendor whose main product is x86 processors.

The house of Zen on Tuesday announced its Q3 2026 results, which saw it grow revenue 45 percent year-over-year to $13.4 billion. The company's Data Center segment won $8.5 billion of revenue, up 97 percent year-over-year. Sales of silicon in AMD's client and gaming segment slid eight percent to $3.7 billion, a worse fall than the five percent fall that helped to sink Intel's share price.

Intel, AMD's great rival in the x86 market, offered revenue guidance of $15 billion to $16 billion for its next quarter, which would represent almost no growth.

AMD, by contrast, told investors it expects around $17 billion of revenue next quarter.

CEO Lisa Su predicted even better days are ahead.

"We think 2027 is going to be a very strong momentum year for both the CPU business as well as the data center AI business," she said on the company's earnings call. "We have said the overall company will grow greater than 80 percent, and for the data center CPU business, we will about double, and for the data center AI business, it will significantly more than double."

Su said AMD is confident its supply chain can meet that demand.

"We have worked across our supply chain to secure the wafers, substrates, advanced packaging, memory, and other key components needed to support much higher growth next year," she said. "We are also deepening our collaboration with our supply chain partners and investing alongside them to secure the long-term capacity our growth will require."

The CEO added that AMD has "strong confidence in our ability to sustain exceptional revenue growth for years to come, with earnings growing even faster."

If those predictions are correct, AMD will likely accelerate away from Intel to become the leading provider of x86 processors after years of challenging Chipzilla's lead. Intel would still have higher overall revenue, due to its foundry operations, but in terms of revenue earned from selling chips, AMD would become the market leader.

AMD is arguably already the server CPU technology leader, and while it is second to Nvidia in the GPU market has trounced Intel.

Even AMD's declining client and gaming revenue has an upside because it was the gaming part of the portfolio that created the problem. PC chip sales rose seven percent year-over-year to $2.9 billion. Su said strong sales of Ryzen mobile processors drove that growth.

"Commercial is becoming a larger part of our client business as large enterprises adopt Ryzen PRO PCs across their employee base," she said, pointing to strong adoption at the world's largest banks, automotive companies, and healthcare concerns. Big corporate buyers are often a source of bigger profits and/or steadier revenue, so AMD looks to be taking the cream off the top of the PC market rather than scrounging for SMB and retail scraps.

Investors liked what they heard, sending AMD's share price up by around five percent in after hours trading. ®

Australian design software maker Canva is restructuring, with 1,271 employees, or 22 percent of its workforce, facing layoffs, changes to their roles, or requests to relocate.

The company, which has become a challenger to Adobe, announced the news by sharing a message that CEO Melanie Perkins sent to staff on Wednesday.

That missive, titled "An important update on our team," doesn't mention AI but does state "New technology is transforming what people can create, how they create it, and what they expect from the products they use."

Perkins observed that Canva's rate of growth recently slowed for the first time in a decade and shared her view that continued success won't happen without change.

"After exploring a wide range of alternatives, I've come to the difficult conclusion that small changes won't be enough," she wrote. "To set ourselves up for the next decade, we need to refound Canva, getting back to our roots of small, empowered teams with the freedom to make decisions, take ownership and focus on what matters most for our community."

The CEO did not say how many jobs would go, but argued the changes are necessary because the company has become inefficient.

The changes will therefore see Canva combine its product, growth and platform teams into a single "Product Org," which Perkins says means "changing the way we work to provide clearer ownership, greater autonomy and flexibility to tackle our most important goals."

She characterized the changes as "simplifying layers, consolidating overlapping responsibilities, and ensuring teams have the right mix of skills and experience for the chapter ahead."

The CEO three times described herself as "deeply sorry" for the changes.

Affected staff will receive a calendar invite for an optional individual meeting with a senior leader. Remaining staff get to attend two rounds of meetings next week - one company-wide affair and another with their reorganized teams.

Another of Perkins' decisions will see Canva move more of its product teams to its headquarters in Sydney, Australia. The company will ask some staff to relocate and has promised "comprehensive support" to make that happen.

Departing staff can keep their company-issued MacBooks and will receive at least six months' pay, a year of health insurance, plus various forms of support to find a new gig including a one-year membership of LinkedIn Premium.

Perkins' post does not cite financial pressures as a reason for the layoffs. Canva is privately held and does not face the same financial disclosure requirements as a listed business, but last year revealed it had 260 million monthly users and revenue of $3.5 billion. Execs have previously claimed the company is profitable on a free-cash-flow basis.

Canva and Atlassian are among Australia's most successful and celebrated technology businesses. Both have now announced job cuts this year. ®

North Korean cryptocurrency thieves tricked a Binance software engineer into downloading malware onto their work-issued computer during a fake job interview process before last week's record-breaking $6.3 billion heist.

In an incident report published on Tuesday, the cryptocurrency exchange said several techniques - including social engineering via fake job interviews, malware downloads, compromised internal AI tooling, endpoint takeover, and an "advanced transaction-authorization method" - led to the October 29 security breach.

"Our current investigation indicates that this was a sophisticated, multi-stage campaign," the incident report said. "It was not a single isolated breach or a simple phishing incident."

Over the weekend, the FBI attributed the theft to TraderTraitor, a North Korean hacking operation linked to Lazarus Group, known for combining espionage with massive cryptocurrency heists. The Pyongyang-backed crew is also highly adept at job market scams - both posing as IT workers, who are then hired by Western companies, and as potential employers to lure software engineers into downloading malware and infecting corporate devices.

With the recent Binance attack, the Norks used the latter.

The attack prep began back in June with a supply chain attack when one of its engineers downloaded a malicious Homebrew installation. While the malware was detected by the security team, the device was replaced, and all affected credentials rotated, "Engineer A's role and his related colleagues on our wallet infrastructure team was exposed to the hacker," the report says.

Three months later, the social engineering part of the attack happened.

In September, another software engineer was contacted via a "legitimate job-hunting platform" by someone purporting to be a project director who arranged a series of job interviews for the Binance engineer with the fake company. During the first round - a live video interview with the engineer and a fake HR rep - the interviewer "shared a social media account with him, including personal photos and videos posted on the platform, in order to build trust," according to the report. "This was more advanced and deceptive than previously seen social engineering approaches."

The second interview with someone purporting to be the CEO of the phony company was conducted on a "popular social chat app," and toward the end, the CEO instructed the engineer to download an "assignment" containing a malware-laced code repository. The code executed on his work computer during the setup process, which kicked off a series of unauthorized activities.

According to Binance, the malware replaced a legitimate development tool component with a malicious version, which executed commands and established communication with an attacker-controlled external server. The attackers also took steps to hide the malicious script and evade detection, and it worked. Binance's endpoint security tools did not immediately detect the malware.

Here's the script:

CREATE /opt/homebrew/Cellar/pyenv/2.8.0/libexec/pyenv_

MODIFY pyenv_ → 3,414 B

DELETE .../libexec/pyenv (3,344 B) ← remove the valid pyenv

RENAME pyenv_ → pyenv (3,414 B) ← replace it with a malicious version

DELETE /opt/homebrew/bin/pyenv (3,414 B)

ln (create a new symlink) + spawn zsh

Then, on October 20, the miscreants used the compromised computer and an internal AI tool to gain access to the computers of wallet DevOps engineer A and another wallet software developer, Engineer C.

Engineer A had a container deployment permission token on his computer, which the Norks soon discovered. On October 29, the attacker accessed Engineer C's computer, used the stolen credentials from Engineer A to deploy a container, and triggered a script to launch the cryptocurrency-stealing attack.

As of Tuesday, Binance said it has rebuilt the compromised wallet system and internal AI tool, and replaced all of the affected devices. It also decommissioned all credentials and tokens assigned to the compromised engineers, and is working with law enforcement, blockchain intel firms, and security researchers to trace and freeze the assets.

The crypto exchange has also hired independent cybersecurity experts to conduct a cybersecurity review of the incident, and recommend additional protective measures. Binance said it's also working to establish a security research fund and initiatives with AI companies and security shops to focus on AI security in blockchain environments.®

With RAM getting more expensive, Microsoft is putting more effort into making Windows 11 run better on systems with just 8 GB of memory.

Microsoft said on Tuesday that it was "raising the bar for 8 GB Windows 11 PCs," following a renewed focus on memory optimization earlier this year. The changes come as Redmond continues pushing more expensive Copilot+ PCs, which require at least 16 GB of RAM, amid a memory crunch fueled by AI infrastructure demand.

Now, Microsoft says a low-end machine with its 8 GB of precious RAM will be up to 20 percent more responsive and 40 percent faster at multitasking thanks to improvements made in memory management.

The tweaks are relatively simple, as Microsoft explained, and seem more about adjusting back to the reality that its flagship operating system needs to function on the hardware people can actually afford to purchase if and when they're forced to upgrade.

Windows 11 will now keep up to 2 GB more active data in memory before shunting it to slower storage, as one example. In another instance, Windows will connect directly to Edge when the browser is already running in the background, allowing it to launch up to 33 percent faster, according to Microsoft. Microsoft apps are also being made more efficient, with Teams able to enable Efficiency Mode by default on 8 GB machines, which Microsoft said will make it launch "up to 20 percent faster."

Memory efficiency is also being improved by tweaking things like Widgets, which will only load when used instead of always being on, and decreasing the time when Edge tabs go to sleep from two hours to 15 minutes by default. Microsoft said it's enabling third-party apps to make use of those tweaks, too.

Microsoft also included benchmark data in the blog post from a recent bit of research it commissioned. The benchmark report compares a range of 2026 Windows machines with models from 2021, finding that a 2026 Dell XPS 13 with 8 GB of RAM performed about as well as a 2021 Dell XPS 13 with 16 GB across selected workloads.

What that means, says Microsoft, is that "the complete PC, not memory capacity alone, shapes the experience."

In other words, please buy some new Windows hardware, even if it's expensive - we're begging you. Software optimization can only do so much.

"For many people, 8 GB is a great fit for the things they do every day," Microsoft EVP of Windows and devices Pavan Davuluri wrote in the blog post. "Demanding workloads, such as 3D modeling, advanced video editing, or large-scale data processing benefit from 16 GB of memory or more."

It's no surprise Microsoft is improving the performance of lower end systems at the same time it's subtly asking users to buy newer kit, even if it's just an 8 GB machine - Microsoft may have done well in the past two quarters, but it's no thanks to Windows devices.

Microsoft's Q4 2026 earnings in late July saw the company rake in cash more broadly, while simultaneously reporting a 7 percent decrease in Windows OEM and devices revenue.

Likewise, in its fiscal Q1 2027 results, reported just last week, Microsoft again brought home a mountain of net income, with adjusted earnings reaching $36.4 billion, up 18 percent from the same period last year. Windows OEM and devices revenue, however, was down a whopping 22 percent. It's hard to think of a reason beyond the fact that hardware prices have shot up with more force than a SpaceX rocket - memory prices are likely to actually make it to Mars, unlike Starship - leaving both consumers and enterprises to stretch those upgrade cycles.

If you need to get the most out of that 8 GB Windows 11 PC, you'll find the latest improvements in 26H2, build 26300.9818 or later. Microsoft is also encouraging users to keep its branded apps up to date, as "some improvements arrive through app updates or through Microsoft Store." ®

Nvidia doesn't seem to care whether open or closed models win out as long as its hardware is powering the infrastructure.

"We love closed models," CEO Jensen Huang said. "There's a solid place in the world for closed frontier models. But the world has so many good ideas … The idea that all of us would be based on general intelligence is not logical."

Speaking with Lin Qiao, CEO and founder of Fireworks.ai, at the company's Forge event in San Francisco on Tuesday, Huang wasn't ready to throw Nvidia's cloud customers, which help deliver the proprietary AI of Anthropic and OpenAI, under the onrushing open model bus.

But having come to the defense of open weight models in July to push back against protectionist pressure from the leadership of Anthropic and OpenAI, Huang recognizes the economic and strategic appeal of AI services that are more affordable and less locked-in.

Most of the value in the world, he said, is captured in specialized intelligence, which is to say AI models trained on specific domains of knowledge.

"The opportunity for artificial superintelligence is already here," he said. "By taking an open frontier model, adapting it, and providing it with your domain knowledge, your access to your specialized tools, and the proper evaluation, you should be able to create essentially a superintelligence."

That aligns with Fireworks' business - offering a platform for companies to train and run models tuned to their specific needs. Coincidentally, that platform just received an infusion of 25,000 Nvidia GPUs to support its expansion in Europe.

Huang's endorsement of open models also fits with Qiao's introductory remarks about the need for the AI frontier to be open. But he largely avoided echoing Qiao's focus on the cost savings allowed by working with open models.

In her introductory remarks, Qiao stood on stage backlit by a slide composed of ten headlines highlighting the extraordinary cost of unchecked token spending.

"Cost is out of control," she said. "Business cannot scale without borrowing a lot of money."

That theme was echoed in a subsequent session helmed by Kay Zhu, co-founder and CTO of AI workspace Genspark, titled Gen-1 Slides: Frontier Quality Without Frontier Costs. Gen-1 Slides, Genspark's model for generating slide decks, was built atop MiniMax M3 and tuned to deliver results comparable to Anthropic's Opus 5 at 1/17th the cost - $0.30/M tokens compared to $5/M.

Asked what might become commonplace next year but seems unlikely today, Huang said this year's breakthrough has been agent harnesses that turn AI models into agents. He added, "I would say the big breakthrough we'll talk about next year, maybe this is a little self-serving, is how Open Shell has enabled every company in the world to deploy agents safely, confidently."

Given the nod to make a second prediction, Huang added, "I would say that the second big breakthrough for industry would be hardware in the loop. And hardware in the loop basically is robotic systems."

Given Nvidia's evident interest and investment in robotics, that would be fortuitous. ®

Revolut says that in less than a year it's managed to slash its AI coding inference costs by roughly 90 percent by making self-hosted models its default instead of relying primarily on commercial services like Claude.

Speaking at the AI Engineer conference in New York last month, David Abitbol, ops and infrastructure lead at the British fintech, described how it saw a 60 percent improvement in coding time and a 70 percent improvement in lead time following its rollout of AI coding tools to roughly 2,000 engineers. Revolut initially adopted GLM-5.1 before moving to GLM-5.3, which it now serves using 56 Nvidia B200 GPUs.

Prior to its switch to GLM, Revolut had been experimenting with AI, deploying Gemini in early 2025 and Claude Code in early 2026. "We were highly skeptical about the cost of AI. We were seeing engineers spend $1,000 a month, $10,000 a month with not so much ROI, and so when we launched it we launched it with strict budget limits," Abitbol said.

Because of this, engineers only had access to a limited number of tokens, so the company began investigating self-hosting iyd own models. At $0.30 per million input/output tokens, self hosting GLM 5.1 cost about a tenth of using OpenAI's Codex or Anthropic's Claude Code. And, as it turned out, when tokens are cheap, you stop worrying about how many they use or what they use them for.

While Revolut did see improvements to coding speed, Abitbol noted the biggest bottleneck wasn't actually the code. Instead, he says the biggest gains came from letting the AI models deal with things like debugging and Jira and Confluence tickets. "When you have unlimited tokens, you're free to do that," he said. "You no longer have to save your tokens for the end of the quarter or the most difficult task."

Developed by Chinese open weights model dev Z.ai, GLM 5.1 isn't exactly a small model. At roughly 754 billion total parameters (40 billion activated for each token generated), the model requires roughly 380 GB of GPU memory just to hold the weights and another 6-26 GB per user to maintain context depending on precision.

To serve GLM-5.3 at scale, Revolut deployed 56 Nvidia B200 GPUs, equivalent in GPU count to seven eight-GPU DGX B200 systems. But as Abitbol noted, the greatest performance gains didn't come from adding GPUs, but rather tuning the model's hyperparameters along with switching to NVFP4 quantization, implementing disaggregated prefill and decode, and tuning context lengths.

One key takeaway was that supporting a one-million-token context window doesn't necessarily make such a large window useful or efficient. "Long context is a scam. You really need to be mindful of that," Abitbol said. "Even though you can serve one million tokens, 256 [thousand] is really what you need."

With these optimizations, Abitbol says that the company was able to achieve sub 3-second time to first token on context windows of around 256,000 tokens, and generation speeds of around 200 tokens a second, both of which he notes are the sweet spot for the company's software engineers.

"What you want to aim for is a sub-3 second time to first token. This is empirical evidence. As soon as you exceed 5 seconds, my Slack inbox starts to blow up," Abitbol said. "In terms of throughput, it doesn't really matter as much. We serve, in terms of interactivity, 200 tokens per second. People are fine [with that]. You don't need to go higher. The way people work has changed. We no longer look at the stream of thinking tokens. People run tasks in parallel and check in when the task is done."

And if throughput is a concern, he notes engineers who sign in after 5 pm are getting closer to 800 tokens a second.

However, it's worth noting that GLM hasn't replaced GPT or Claude entirely. The Chinese model is text only, so it isn't ideal for things like UI work. So instead, Revolut used a model router to direct vision queries to models like Gemini or GPT Luna while keeping code generation running on GLM.

"Open source is now a priority at Revolut, and our economics are favorable. The billing is flat, we don't have any bill shock at the end of the month, and the frontier is the exception," Abitbol said.

Having said that, Abitbol acknowledges that Chinese open weights models may be a concern for privacy and security reasons, but emphasizes that the major concern there is accessing them via APIs running out of China. When run on-prem or on regional cloud resources, there's really nothing to worry about. ®

The Center for a New American Security posits that the next big threat from our future AI overlords involves "sovereign agents" acting on their own and regularly improving themselves.

These are AI agents that operate outside of human control and "pursue open-ended, evolving survival," AI policy researcher Ben Hayum said in a Tuesday report from the bipartisan national security think tank, which was founded by Washington insiders with experience in diplomacy and military affairs. "If sovereign agents are not already possible, they will be soon."

Agents that can maintain the model and compute powering them and independently evolve their goals over time have a "minimal form of sovereignty," Hayum wrote.

A more advanced sovereign agent, however, could also acquire resources to pay for ongoing needs (like compute), grow a distributed population of agents and coordinate them, withstand efforts to shut down, and self-improve.

AI models could either steal or buy compute and model access, Hayum argues, pointing to Anthropic's claims that small AI models could replicate onto IoT devices like smart TVs, and OpenAI's agents' password and access key theft during the Hugging Face hack.

"If sovereign agents accumulate resilient, decentralized, and untraceable infrastructure, they could trigger an irreversible loss of control," according to the report. "In other words, proliferation with no kill switch. If sovereign agents became malicious, they could attempt to wreak havoc from cyberspace, attacking civilians, critical infrastructure, and national security systems alike."

Dun dun duuuun!

There is still hope for civilization as we know it, as long as policy makers and AI developers "respond with bold pragmatism" and enact policy recommendations laid out in the report.

First, frontier labs should name, evaluate, and protect against sovereign agents in their safety frameworks - and lawmakers should pass legislation to require transparency and AI incident reporting.

The report also recommends AI developers, working with the National Security Agency and other federal agencies, build airgapped testbeds to study sovereign agents and kill switches to control them. And it suggests a public-private partnership modeled on the Conficker Working Group - formed in 2009 to combat the Conficker worm - to boost chokepoints against sovereign agents before they emerge.

Other policy recommendations suggest lawmakers enforce human verification and cybersecurity standards for services that the working group identifies as essential to sovereign agents, and calls on departments of State, Treasury, Justice, and Homeland Security to disrupt any gray-market access to unauthorized services.

Finally, "Congress should hold those who release sovereign agents liable for the harms that follow," the report recommends.

This hits on something your humble Reg vultures have been hissing about for quite some time now - and would likely solve all future incidents involving rogue, or sovereign, AI agents: hold frontier labs' human bosses legally liable for their agents' illegal actions. ®

The GSMA wants governments and regulators to work more closely with mobile operators to help protect networks against growing risks from extreme weather.

As the global body representing the mobile network industry, the GSMA claims that severe storms and floods already cost operators $1.7 billion annually to repair damage to cell radio sites, and it believes that figure could nearly double to more than $3 billion a year by 2050, under the climate conditions expected then.

The GSMA sets out its findings in a six-part research series, Futureproofing Connectivity: Climate Risk and Resilience in Telecommunications, alongside a separate document outlining its Policy Principles for Mobile Network Resilience.

The report identifies weather-related risks that can affect telecoms networks, along with the associated restoration costs. Towers, masts, and rooftop sites are particularly exposed to wind and storms or lightning, for example, while ground-level equipment at mobile sites can be inundated in flooding incidents.

High temperatures can cause radio equipment to overheat, while heavy rain can degrade wireless signal strength, and infrastructure in coastal locations may be exposed to storm surges.

However, the report fingers the electricity supply as the biggest vulnerability. If that gets knocked out, the network could go offline once any backup power is exhausted, which may take anywhere from a few hours to considerably longer, depending on the site.

Under current climate conditions, the cost of damage to cell sites is estimated at $1 billion associated with severe storms, while river and coastal floods account for $577 million and $124 million, respectively.

Based on projected mid-century climate conditions, the GSMA expects costs due to severe storms to rise significantly to about $2.37 billion, accounting for three-quarters of all damage.

The report asserts that cell network resilience requires a whole ecosystem approach, pointing out that if severe weather impacts power networks, then that also affects the telecoms infrastructure.

The GSMA says that network firms have traditionally managed the operational risks that may affect network availability, but claims they face barriers in conducting full assessments of those associated with climate change.

Some of these barriers are internal, such as lack of shared information between departments. Other barriers are in the different regulatory reporting frameworks, with infrastructure rules tending to focus on service continuity, while climate disclosure frameworks focus on governance, financial exposure, and reporting for investors and other stakeholders.

The report lists separate recommendations for network firms and for policymakers and regulators.

Operators are advised to embed resilience by design, using modeled projections for severe weather and future risk scenarios. They are also told to plan and test resilience across the wider ecosystem, engaging with others such as public authorities.

When it comes to policymakers and regulators, the report asks for resilience frameworks to be developed through consultation with the operators. It also advises coordination among mobile operators, electricity and fuel providers, and emergency services to help maintain vital communications during major disruptions.

And as if that isn't enough to chew over, the separate policy document sets out five principles for governments and regulators. These include avoiding prescriptive regulations and instead promoting outcome-based policies, encouraging dialog and consultation between governments, regulators, and operators, and developing resilience as a cross-sector responsibility.

"Mobile connectivity is essential infrastructure," says GSMA Foundation president and chief regulatory officer John Giusti.

"Whether responding to extreme weather, power failures or other major disruptions, people, businesses and emergency services depend heavily on mobile networks," he added.

"Severe weather is already costing the industry billions of dollars, and those costs are set to rise substantially. Building resilience will require a coordinated effort amongst operators, governments and the wider ecosystem." ®

Researchers at Sandia National Laboratories have developed a method to make asteroid deflection modeling faster and more detailed. Their simulations also demonstrated just how much asteroid shape and orientation can complicate efforts to deflect a threatening space rock.

Sandia researchers, led by staff physicist Nathan Moore, were looking for a way to better model how X-rays from a nuclear detonation near an asteroid can change its trajectory, as explained in a recently published research paper.

"Before, the calculations were so time-consuming, one could not do very many of them," Moore told The Register. "So it was very difficult to explore different scenarios, different asteroids, or different mission concepts."

The complexity of those calculations forced earlier researchers to make compromises, often modeling asteroids as spheres or simplifying the physics involved.

Perfectly spherical asteroids are rare, and their irregular shapes present an additional challenge for those keeping an eye on the skies to avert a potential asteroid apocalypse.

To overcome the computational problem and enable better modeling of irregular asteroid shapes, Moore and his team combined two high-fidelity simulation codes already in use at Sandia: ALEGRA, which models radiation heating and high-energy material dynamics, and SPARTA, which simulates the behavior of low-density gases.

As Moore explained to us in an email, both codes are good at one particular part of asteroid impact modeling, but bad at the other.

"ALEGRA … can very accurately simulate the vaporization of the rocky surface, but the calculation becomes more inefficient for matter at low densities," Moore told us. "This is a problem for asteroid deflection simulations, because as the gas expands around the asteroid, the calculation time goes up."

Earlier asteroid simulations using ALEGRA typically modeled asteroids as spheres to reduce calculation times.

"SPARTA is more efficient at low gas densities," Moore said. "The problem is, conversely, SPARTA gets more difficult to run at high gas densities. So the key here was using both codes."

In Moore's model, ALEGRA handles regions of high gas density, while SPARTA tackles the more dilute regions of an irregular shape. The result is two codes handling the parts of the problem they're best suited to solve, reducing the overall computational burden.

The combined approach is more than 1,600 times faster than earlier methods, Moore said in a Sandia press release, while retaining sufficient accuracy to predict deflection in irregularly shaped asteroids.

The improved simulations reveal just how much asteroid shape and orientation can complicate attempts to calculate the outcome of a nuclear deflection.

"We expected aim to matter," Moore said. "What surprised us was how much."

Deflection simulations using detailed models of three well-studied asteroids - Golevka, Bennu, and Itokawa - found that irradiating different sides of the same asteroid with X-rays from a nearby nuclear detonation could produce as much as a fivefold difference in the resulting velocity change. Bennu, the most symmetrical of the three, still showed nearly twofold differences in deflection velocity depending on which side faced the blast.

In other words, we need to be incredibly precise about where we detonate a nuclear device near an asteroid, or we risk giving it a much smaller shove than intended.

With that in mind, the paper points out that we might know little about an incoming asteroid, and might not have enough time to get a good picture of it before attempting to deflect it and prevent life as we know it from ending in a planet-wide conflagration. We also can't be quite sure an asteroid's orientation will cooperate once our deflecto-bomb arrives. The team also pointed out that it's tricky to get an accurate prediction without knowing the composition and internal structure of an asteroid.

One small step for asteroid deflection, one giant heap of new things to worry about, in other words. But hey - at least we're no longer limited to deflection models that treat asteroids as perfect spheres, even though the real threats are likely to be considerably lumpier.

Let's just hope we spot an asteroid threat from far enough away that we get a chance to study it first. ®

Password manager mSecure has warned customers to change their passwords, reset two-factor authentication, and potentially remove their passkeys after detecting suspicious activity, but it has not explained what happened or whether anyone's secrets were exposed.

In an email sent to customers on November 3, seen by The Register, US-based developer mSeven Software said it launched an investigation after discovering something amiss with its password management application.

"mSeven Software ('mSeven') recently learned of suspicious activity associated with the mSecure password management application," the company wrote. "In response, mSeven quickly launched an investigation, took steps to mitigate the issue, and engaged external cybersecurity experts to assist."

The company has yet to answer what the suspicious activity involved, when it occurred, or whether attackers accessed customer information.

Its advice to customers, however, suggests that they should take this seriously.

"Given the potential security implications, mSeven recommends that you immediately take steps to protect all confidential information stored within the mSecure password management application," the email continued. "This may include changing passwords, resetting or re-enrolling two-factor authentication methods, or removing existing passkeys."

For anyone with hundreds of passwords stored in mSecure, that's potentially hours of work. And without knowing what happened, customers have no way of telling which credentials they actually need to replace.

The warning has understandably gone down poorly with customers, who have taken to mSecure's support forum demanding answers.

"Please provide more context to your email below advising there has been a security breach," one customer wrote. "It is not acceptable to offer the vague advice 'take steps'. Please state what remedial action is required, and what data was compromised."

Others are wondering whether they're expected to spend the next several days changing every password they've ever saved.

"I have over 800 passwords in mSecure, is the email saying I need to change all of them!" another customer asked.

A separate user, who said they had more than 500 passwords stored in the application, questioned whether the warning applied to every two-factor authentication method and passkey associated with their accounts, or just those used to access mSecure itself.

The confusion has also spilled onto Reddit, where customers have shared copies of the vague notification and questioned what might have happened.

mSeven has yet to respond to The Register's questions, leaving customers to guess whether their data has actually been compromised or the company is simply being cautious. That's not exactly reassuring from a business whose entire purpose is keeping other people's secrets safe.

The company's email concludes: "We regret any inconvenience that this may cause." For customers contemplating hundreds of password resets, "inconvenience" may be putting it mildly. ®

GhostBSD, the easiest-to-install flavor of FreeBSD for the daemon-curious desktop dabbler, is back with a fresh update.

GhostBSD 26.2-R15.1p6, to give it its slightly unwieldy full version number, is based on FreeBSD 15.1. It brings a new installer, full-disk encryption support, a smaller installation image, and fixes across its desktop tools.

FreeBSD 15.1 was released back in June, and at the time of writing, it's the most recent point release of FreeBSD.

The original plan was that FreeBSD 15.1 would include the option of installing KDE Plasma directly from the setup program, so that the OS would have a graphical desktop from its very first boot. However, this feature isn't ready just yet.

That gap inspired Daemonarchy, which we covered late last month. But its Omarchy-derived, keyboard-driven tiling desktop can feel unfamiliar even to experienced Linux users accustomed to conventional desktops.

GhostBSD has different goals. It brings Linux-like desktop convenience to FreeBSD: a live graphical environment to explore before installation, a graphical installer, and tools for managing software and updates.

These conveniences are familiar to desktop Linux users, but a standard FreeBSD installation takes more work. Its text-mode installer leaves users to install and configure a graphical desktop themselves, if they want one.

This is just how the BSD family does things, and most BSD users we've talked to regard it as standard operating procedure and see nothing wrong with it. For Linux users who are curious about how things are on the other side of the FOSS Unix fence, though, those extra steps can form formidable barriers. GhostBSD is an attempt to lay down a welcome mat.

The standard edition of GhostBSD installs the MATE desktop, and the downloads page also offers a community-maintained version with Xfce instead. As the project said it would back in February, this release uses the XLibre fork of the X.org X11 server.

That's one fewer desktop than when we last tried GhostBSD in August 2025. That release also offered the Mac OS X-like Gershwin environment.

There are practical hardware fixes too. The release notes list improvements for Intel AX200 Wi-Fi cards, open networks, password alerts, and non-ASCII network names, while USB HID support is enabled by default.

Gershwin remains in active development, so we asked project lead Joe Maloney why it wasn't included. "It has been difficult to find time to package for GhostBSD. It's the only distro I have ever packaged for," Maloney said. He added that rapid contributions from AppImage creator Simon Peter, known as probono, had made packaging increasingly difficult.

He also pointed to the fact that GhostBSD project lead Eric Turgeon is developing his own environment too: the Mocka desktop. Mocka is still in its very early stages - alpha version 0.0.2 appeared early last month - but the plan is that when it's ready, it will become GhostBSD's default desktop. For now, the only available module is a dock for the MATE desktop, but the curious can follow its development on the GitHub project page.

There may be no Gershwin included, but GhostBSD 26.2 offers MATE 1.28, and there's also a community-supported version with Xfce 4.20 instead. We tried the default MATE edition, and it works as well as ever. It defaults to a dark theme and the classic two-panel GNOME 2 desktop layout. The System menu in the top panel has an Administration submenu with a few of the project's own apps that aren't from upstream FreeBSD: "BE Station" for managing system boot environments, "Software Station" for browsing and installing applications, and "Update Station" for updates.

We regularly hear from users growing tired of the Linux vendors' headlong rush into Rust, Wayland, systemd, containerized app formats, and so on. Some are curious about whether FreeBSD offers a usable alternative. GhostBSD is the easiest way to get a complete, ready-to-use graphical FreeBSD desktop. There is a price to pay for switching to FreeBSD: for a start, most proprietary Linux apps, from Chrome to Slack, are not available here. Hardware compatibility needs more checking than on mainstream Linux, especially with recent GPUs and Wi-Fi adapters. If Windows gaming is a priority, this isn't the obvious choice.

But if you can live without those proprietary apps and find suitable open source replacements, FreeBSD is a viable alternative - and GhostBSD takes much of the setup work out of trying it. ®

A phishing attack on Harvard University last year netted the details of more than 850,000 individuals, including students, alumni, parents and family members, faculty members, and donors, according to data breach service Have I Been Pwned (HIBP).

The Ivy League institution first confirmed the attack, later claimed by ShinyHunters, in November 2025, although HIBP gave the first indication about how many people may have been affected by the breach for the first time today.

Run by venerable infoseccer Troy Hunt, HIBP ingested all the files and revealed that the types of data stolen from the university extended beyond the personal and donation information it referenced in its initial disclosure.

According to HIBP, the 850,000+ figure pertains to unique email addresses found in the data dump. The following data points may not have been collected about each and every individual, but are included in the stolen files:

• Names

• Dates of birth

• Places of birth

• Ethnicities

• Genders

• Marital statuses

• Occupations

• Job titles

• Email addresses

• Physical addresses

• Phone numbers

• Salutations

• Social media profiles

• Citizenship statuses

• Education levels

• Family members' names

• Employers

• Charitable donations

Harvard did not immediately respond to our requests for more information.

Its disclosure at the time described the attack as an instance of "phone-based phishing" that targeted its Alumni Affairs and Development department, leading to unauthorized access to its systems.

It did not specify the attacker by name, although ShinyHunters claimed responsibility for the intrusion on the same day it listed the University of Pennsylvania on its leak site.

The group claimed to have stolen more than 1 million records from Harvard and 1.2 million from Penn.

In Penn's case, HIBP pegged the total number of affected individuals at around 623,800, based on unique email addresses, although additional data types such as donations, personally identifiable information, religions, and more were affected.

ShinyHunters' raid on Harvard systems came around a month after the university probed a separate attack by Cl0p, which at the time was on a tear exploiting a zero-day in Oracle's E-Business Suite.

The impact from that attack was limited to a small amount of data at a small administrative unit, however.

Earlier this year, Harvard was also affected by ShinyHunters for the second time in the space of a few months after the cybercriminals attacked Instructure, the EdTech vendor behind learning platform Canvas.

The university pulled its Canvas site offline while Instructure handled the cleanup, although it never officially commented on whether any data was compromised as a result. ®

Most JavaScript programmers default to their ride-or-die frameworks, but such loyalties can hinder them when they task AI agents with doing their dirty work. Human-friendly tools aren't necessarily the most effective ones for the model-powered minions.

"One of my pet peeves with the web development community is that so many of the decisions we make are vibes-based," wrote Seattle-based Nolan Lawson, currently a developer at AI security firm Socket. That's an understatement; the JavaScript wars have been well underway for over a decade now, with each new framework promising to cure the woes of last year's release.

Not wanting these vibey biases to seep into the realm of AI, Lawson created a benchmark to compare how different JavaScript frameworks perform under the same agentic model. He subsequently ran the test on 12 JavaScript frameworks, publishing the results in a post that compared each framework in terms of build cost, maintainability, and end-user performance of their agent-built apps.

The results are illuminating. In general, agents tend to work more efficiently with server-centric frameworks than with client-heavy "Single-Page App" (SPA) frameworks (which should tickle for React haters).

But the biggest take-away should be less controversial: agents best utilize those frameworks that favor convention, API stability, and concision.

For the experiment, Lawson wrote a set of instructions for a Claude Code agent to build a basic Pokémon web application using the PokéAPI. The model used the identical setup on 12 different frameworks and JavaScript stacks: Angular, Astro, Gracile with Lit, Hono with htmx, vanilla JavaScript with Vite, Next.js, Nuxt, Qwik, React Router, TanStack Start, SolidStart, and SvelteKit.

Lawson evaluated each stack across five independent sandbox runs; the repeated iterations smoothed the variance in the model's performance.

To verify the resulting app met the specs, Lawson deployed a custom automated test suite that queried the browser's accessibility tree to ensure interactive elements like buttons and inputs weren't faked - a shortcut LLMs occasionally take. Lawson also used Google Lighthouse to measure transfer size and runtime performance.

If the frameworks specified official MCP servers, Claude plug-ins, and agent-skill files, Lawson provided them as resources to the agents (though they weren't always consulted). Otherwise, he did not mandate any specific architecture for rendering the app.

The differences between the most cost-effective builds and the most expensive ones are substantial, to say the least, with the best lapping the efforts of the stragglers.

If you want to build a Pokéworld cost-effectively, the strongest bet is on Hono paired with htmx. This tag-team of technologies had the most economical run, with a median cost of $2.67 per Pokéapp.

Perhaps this is not surprising, given that the duo runs entirely on the server, thereby eliminating a whole set of checks required to synchronize an architectural split between client and server. It also helped minimize the code, which resulted in fewer output tokens being consumed.

Nearly as effective was Astro ($2.76). Like Hono/htmx, Astro does everything server-side. Server-side data crunching also helped Astro achieve the fastest median time to build and test the app, at 8.2 minutes.

Nuxt ($2.82) was the most cost-effective option for an SPA-like approach. Nuxt renders code on the server that gets hydrated on the client. Nuxt is actually built on Vue.js, but is "opinionated," eliminating a lot of boilerplate that the agent must write. Nuxt also handles data transfer and file routing internally, which is more code not to fuss over.

It's also worth noting Nuxt consumed the fewest tokens of the bunch (3.85 million), though that did not make it the cheapest, given how Claude charges different rates for input and output tokens. Nuxt produced the app in 1,415 lines of code, which is second only to Hono/htmx's 1,370 lines.

Overall, these top three frameworks were within the margin of error of each other, which means any one of them could have gotten the top score depending on what kind of day the LLM was having.

Some of the more traditional frameworks lumbered way behind, however.

A SvelteKit-based app cost $4.28 to generate, with the agent repeatedly going back to the SvelteKit documentation to get a better handle on how the recently-updated Svelte works. Angular was the most expensive way for an agent to build an app, costing $4.97 while taking 17.2 minutes and chewing up 9.76 million tokens (with the agent failing to even finish a working app in two of its five attempts).

All the resulting apps aced the Lighthouse accessibility and performance tests. It's a pretty simple app to render, Lawson acknowledged.

Astro's app had the fastest loading time (called Largest Contentful Paint or LCP) of 810 ms, while the slowest frameworks in the cohort - Next.js (2,090 ms) and TanStack Start (2,028 ms) - took more than twice as long to deliver the goods, perhaps because of all the JavaScript they shipped to the client.

Lawson hopes this work will set the stage for applying a bit more rigor in selecting frameworks for AI jobs, now that developers won't have to do the actual heavy lifting themselves.

But a few developers on Hacker News pointed out AI efficiency isn't the only factor to consider. There's the human component of maintaining the framework as well.

"After a decade of React, I'm not going to go learn htmx or any other marginally better same-shaped tool for a new project, especially when it'd degrade my ability to review and debug another actor's work product," one wrote.

After all, the frameworks themselves might evolve with the help of AI, minimizing their differences over time. "Besides a rewrite to another framework using AI, another option is to modernize the framework with AI instead," someone else noted. ®

Twenty-nine NHS trusts listed as benefiting from Palantir's Federated Data Platform (FDP) said in Freedom of Information responses that they were not using it, according to health worker campaign group Medact.

That represents about 21 percent of the 137 trusts NHS England lists as "realising benefits" as ministers weigh whether to continue the contract.

The government must decide in December whether to continue the £330 million agreement with a Palantir-led consortium supplying the platform.

Medact, which is campaigning against the deal, collected the responses to scrutinize the government's case for retaining Palantir. The relationship between the US tech biz and the NHS began during the pandemic with contracts worth £60 million awarded without competition.

NHS England awarded the FDP contract in November 2023. It says the platform brings together operational data held in separate systems to help staff work more efficiently.

NHS England's September figures list 142 trusts as live on the platform - 137 of which it claims are benefiting.

Among the disputed entries, NHS England cites University Hospitals of North Midlands NHS Trust as receiving benefits from the platform, but in response to an FoI request in July, the trust said: "FDP is not live. We do not intend to roll it out in the future. A variety of factors prevent roll out, but primarily it does not appear to provide functionality over and above what we have already."

Meanwhile, a response from Nottingham University Hospitals said "no implementation timeline has been agreed," according to The Guardian.

The Register asked NHS England to explain the discrepancy.

An NHS spokesperson said: "The NHS Federated Data Platform is universally available for all NHS trusts and integrated care boards across England, with local NHS organisations able to select individual tools depending on their needs.

"137 NHS trusts have now reported benefits from the wide range of tools available on the platform - helping staff join-up and improve care for patients, and work more efficiently."

Palantir maintains its software is playing an important role in improving patient care and helping to deliver 100,000 additional operations, a 12 percent reduction in discharge delays and the removal of 675,000 patients from waiting lists.

In July, the UK's statistical regulator told NHS England to ensure greater accuracy in quoting figures following a row over data used to promote the purported benefits of the FDP.

Dame Chi Onwurah, the chair of the House of Commons Science and Technology Committee, has given the government until today to clarify its response to questions over whether NHS England has any exit plans should it decide to end the Palantir contract in February.

James Frith, a junior minister in the Department of Health and Social Care, wrote [PDF] to the Committee in October explaining that the contract has "measures built in to facilitate exit and transition including clear intellectual property rights in NHS build products or components."

He continued: "Publishing more detailed information on the Exit Plan would, or would be likely to, prejudice commercial interests and NHS England's ability to secure best value for taxpayers in future commercial activity."

Onwurah replied: "I understand that you will not publish the exit plan, but will you confirm if one has been prepared?"

The Register asked Palantir to comment. ®

Britain's AI infrastructure boom could soon have an unlikely new home: Drax Power Station, where the operator of the country's largest biomass power plant has unveiled plans for a £20 billion datacenter development capable of consuming more than a gigawatt of power.

Drax announced on Tuesday that it has teamed up with investment giant Brookfield and its UK-based AI infrastructure business Radiant to build the facility at its North Yorkshire site. The money would come through a joint venture between Brookfield and Qatar's Qai, although the £20 billion headline figure is more ambition than actual money on the table at this stage.

The companies plan to start relatively small, at least by the standards of the AI industry's apparently bottomless appetite for electricity. The first phase would deliver up to 100MW of capacity by early 2028, backed by a proposed £2.5 billion investment. A second phase could add another 500MW between 2028 and 2031, with further expansion taking the site beyond 1GW after that.

However, a few hurdles remain between today's announcement and a functioning AI megafactory. According to Drax's regulatory filing, the first phase still requires planning permission, regulatory and network approvals, commercial agreements, and final investment decisions.

It also needs a tenant, which is a fairly important detail when you're proposing to spend billions on somewhere to house other people's computers.

Drax reckons its existing infrastructure gives it an advantage over developers struggling to secure electricity connections. The 1,000-acre site already has access to 4GW of grid capacity, 2.6GW of active generation and cooling infrastructure, along with proximity to the UK's fiber network.

The UK government estimated last week that all of Britain's existing data centers use around 2GW of electricity. Drax's proposed development alone could eventually need more than 1GW, although how much power it actually consumes will depend on how fully it's used.

The first phase would draw electricity from the grid, but Drax wants to supply up to 500MW directly from its existing generating capacity under a long-term power purchase agreement, potentially allowing AI servers to run on electricity produced by burning imported wood pellets.

That brings another complication. Drax signed a government-backed Contract for Difference agreement last year covering biomass generation between April 2027 and March 2031. The agreement allows the company to request permission to redirect up to 500MW of capacity towards a datacenter, but ministers and the Low Carbon Contracts Company would have to agree, taking into account consumer value, energy security, and sustainability.

The announcement also arrives as enthusiasm for enormous AI server farms is proving rather less universal outside government and industry circles. Opposition to data center developments has been growing on both sides of the Atlantic, with communities raising concerns about electricity bills, water consumption, and the impact of industrial-scale facilities on their surroundings.

A recent investigation found that businesses accounted for 84 percent of the stakeholders UK ministers met to discuss datacenters between July 2024 and June 2026, while environmental groups secured precisely zero meetings.

Drax and its partners are promising a community benefits fund and say public consultation will begin this month. Whether that will reassure residents concerned about another enormous consumer of electricity is a moot point.

Drax CEO Will Gardiner said the project could "help secure the long-term future of the site and protect hundreds of jobs," while supporting Britain's ambitions to become a major player in AI.

The government, unsurprisingly, is enthusiastic. Chief Secretary to the Treasury Emma Reynolds welcomed the proposed investment as an opportunity to create jobs and support Britain's reindustrialization.

"Investments like this will help pave the way to a new age of British reindustrialization, creating good jobs in Yorkshire and delivering growth right across the country," she said.

Ministers may be enthusiastic, but Drax still has to find a customer, secure planning permission, and convince regulators before anyone starts plugging in servers. ®

Virgin Media O2 (VMO2) wants to chop a third of its labor costs and is expected to lay off thousands of workers, giving staff more to think about than Christmas presents during the run up to year-end.

Shrinking the workforce is one of a number of cost reduction measures designed to save the UK's second largest telecoms provider some £800 million by 2029.

It currently employs 16,000 people and redundancies will "equate to roughly 30 percent of the company's total people costs" over the next two years, VMO2 said on Tuesday.

The majority of the redundancies will come from "non-customer facing roles," it claimed. The Register asked the telco to be more specific on the headcount cuts and the type of roles are at risk.

A spokesperson refused to provide a number, saying those plans are being finalized and consultations are due to take place. Logically, if more senior head office positions are hit then fewer people will need to leave to meet the £300 million ($402 million) target figure.

The Reg understands thousands of roles are at risk.

Another lever management at VMO2 is pulling involves halting investment in any further fiber network upgrades beyond existing commitments. Instead, a low-cost upgrade will be deployed to its hybrid fiber-coaxial (HFC) network (i.e. cable customers) to bring 2 Gbps services to the majority of the network in 2027.

VMO2 was formed in 2021 by the merger of broadband and TV biz Virgin Media and mobile operator O2. The pair pledged to spend £10 billion (c $13.4 billion) on beefing up 5G coverage and laying gigabit "full-fiber" fixed-line networks.

Both Virgin Media and O2 talked up those connectivity commitments when justifying the case to combine their businesses, saying they needed greater economies of scale to compete with their rivals.

However, business has not gone as planned, with revenue falling over several successive quarters. The latest figures for Q3 ended September 30 show total revenue down 5.7 percent year-on-year to £2.43 billion. The networks giant blamed consumer fixed market pressures, among other things.

VM02 forecasts a revenue decline of 3 to 5 percent for the whole of 2026.

In the consumer division, VMO2 talked of more clearly defining the role for each of its brands. O2 Business is planning to streamline the product portfolio.

Tom Oughton, senior analyst at Megabuyte, said the changes were expected.

"VMO2 has undergone a strategic review as it looks to address rising leverage, with today confirming a transformation program to ultimately bring c. £800 million ($1bn) in cost savings ahead of its first debt expiry in 2029. The routes to achieve this are the most obvious ones: reduce capex on full fiber and focus on HFC upgrade (lower cost, quick payback) and reduce operational costs by cutting headcount," Oughton said.

"The obvious question is how a significant transformation will impact the business. VMO2 focuses on the benefits, but the business is declining 3-5 percent organically with 16-17k staff, so reducing headcount by 25 percent or more could stretch operations, worsen customer support, and cause broader disruption."

Paolo Pescatore, founder and analyst at PP Foresight didn't mince his words either: "Virgin Media O2 is in serious trouble, with customer losses and declining service revenue and earnings exposing the scale of the challenge."

"The business has lost its strategic direction, and continued customer departures suggest it is struggling to rebuild trust while rivals give people compelling reasons to switch. Cutting costs can buy time, but it cannot repair those relationships or compensate for weak execution," Pescatore added.

"Management must show customers why they should join and stay; otherwise, it risks shrinking the business without fixing what is driving people away. Shareholders need to take a hard look at these businesses and make even tougher choices about their future."

Noel McClean, deputy general secretary of Prospect union, said the job cuts "will be devastating for those directly affected and will of course have an impact on colleagues remaining in the business.

"Industry conditions are tough but this level of cuts will drastically alter the way VMO2 works." ®

The UK's National Cyber Security Centre (NCSC) says AI-powered malware is "highly likely" to be a part of a future ransomware attack on Britain's critical national infrastructure between now and 2028.

The warning accompanies its annual review and a separate assessment exploring how AI could change cyber threats over the next two years.

The cyber arm of signals intelligence agency GCHQ did not provide much detail about what form this AI malware may take. However, it alluded to open source AI security tools that were initially intended to assist security researchers being repurposed for nefarious activity.

Attackers are already repurposing such tools. For example, researchers are increasingly seeing widespread abuse of ARTEX, an open source, Chinese-made AI red-teaming platform, in attacks on UK registrar 123 Reg, Microsoft 365, and others since its launch in July.

In its pessimistic scenario for 2028, the NCSC envisages "a significant increase in the frequency and sophistication of cyberattacks worldwide," with wider adoption of AI tools making attribution harder.

China "has likely played a role" in supplying those tools in this future, "enabling them to hide their activities within the noise," the report states.

It also considers a further threat: an autonomous attacker operating beyond its human operator's control.

The NCSC says it "is a realistic possibility such an entity could self-replicate and conduct cyberattacks at will, with no apparent motivation or restraint."

Such attacks would likely complicate geopolitical relationships, owing again to attribution difficulties, making it easier for countries that order attacks on others to deny their involvement.

At a time when each Patch Tuesday seems to come loaded with record numbers of bugs, the NCSC acknowledges that AI is already changing vulnerability discovery and patching.

The body is also exploring how AI could help defenders keep pace. Project DROMOS began earlier this year with a four-week pilot analyzing government code repositories, before expanding into selected operational environments.

The Government Cyber Coordination Centre (GC3) launched the project with the AI Security Institute and other departments. The NCSC says it uncovered hundreds of potential security issues, including critical vulnerabilities.

The review also reaffirms plans for Cyber Shield, an agentic cyber defense system first teased in July.

Cyber Shield is expected to be delivered in stages over the next five years. It will involve the NCSC providing defensive agents deployed by participating organizations with data that will strengthen national resilience to ongoing cyber threats.

The NCSC will be running the scheme alongside the Department for Digital, Culture, Media and Sport (DCMS), but acknowledged that this kind of automated defensive activity currently sits outside many organizations' risk tolerances, meaning it has a role to play in building trust in the system before adoption can go mainstream.

As reported last week, the government has set aside £60 million to fund the initiative, and an additional £70 million to support a new National Exposure and Real-Time Vulnerability (NERV) Centre.

The NERV Centre's job will be to identify and maintain an ongoing understanding of vulnerabilities across CNI organizations and the wider economy.

Per last week's budget, £40 million will fund a National Agentic Security Unit responsible for responding to uncontrolled bots and other agentic mishaps, and £24 million was assigned to strengthen public sector defenses.

An undisclosed sum will also help establish a new NCSC-managed AI Cyber Security Centre in Manchester to develop AI security expertise across government.

In the meantime, NCSC CEO Dr Richard Horne is calling for organizations to "invest with urgency" in their security and resilience.

"For decades, we have collectively got away with doing tech on the cheap," he said today, announcing the annual report.

"We can be a nation where leaders of all organizations actively choose to invest with urgency... And where all organizations make security and resilience a fundamental requirement, not an afterthought.

"The collective result would be a source of enduring national strength for us all."

Horne said organizations should treat the NCSC's Cyber Essentials certification as the starting point, but scale up resilience efforts in line with organizational size, complexity, and risk.

The government also announced a 75 percent discount on Cyber Essentials certification for micro, small, and medium-sized businesses applying for their first certificate. ®

Apple says it sold more Macs in fiscal 2026 than in any previous year, but soaring memory prices are threatening to spoil the Cupertino-based gadget maker's record-breaking run.

The warning came during Apple's fiscal fourth-quarter earnings call on Monday, the first since John Ternus took over as CEO.

"What we're seeing with memory pricing right now is unprecedented," Ternus told analysts.

Apple has already raised prices across much of its hardware lineup, including Macs and iPads in June and iPhones in September. Those increases, along with existing inventories and cheaper components elsewhere, have softened the blow. Unfortunately for Apple, memory prices haven't finished climbing.

"For the December quarter, we expect even higher memory costs," Ternus said, adding that Apple expects to pay more for memory beyond the end of the year.

Mac buyers haven't been deterred so far. Apple pulled in $10.8 billion from Macs in the September quarter, up 24 percent year-on-year, following $10.35 billion in the June quarter, up nearly 29 percent year-on-year. The MacBook Neo and MacBook Pro accounted for much of the latest growth.

"We have never sold more Macs in a fiscal year than we did this time," Ternus said.

Whether Apple can keep enough of them on shelves is another matter. COO Sabih Khan said the company had already experienced supply constraints across iPhone, Mac, and iPad during the September quarter, largely because of limited availability of advanced chipmaking capacity. Now memory is taking over as the chief headache.

"For the December quarter, we expect constraints on Mac and iPad, primarily driven by memory, as demand has been higher than what we expected," Khan said.

Apple isn't sitting idle. Ternus revealed that its manufacturing purchase obligations jumped by $33 billion between June and September as it bids to secure supplies, although he offered no indication of when the memory crunch might ease.

Higher memory costs are also squeezing Apple's margins. CFO Kevan Parekh said gross margin fell by roughly one percentage point between the June and September quarters, excluding tariff effects, entirely because of memory costs. The damage would have been worse without existing inventory, cheaper non-memory components, and those price hikes.

Apple expects its gross margin to fall from 47.9 percent in the September quarter to between 45 and 46 percent in December, primarily because of higher memory costs. Costs associated with rolling out Siri AI will also weigh on the figure.

Parekh said Apple had "reluctantly" raised hardware prices, but supply constraints made it too early to assess how customers were responding.

Meanwhile, Ternus is keen to sell customers on the Mac as an AI machine, particularly for running models locally.

"Mac mini and Mac Studio in particular are pushing the boundaries of local private AI," he said, pointing to workloads that previously would have been difficult to run on desktop hardware.

Apple has also refreshed its Mac lineup with M6-powered machines and a touchscreen MacBook Pro, giving customers more reasons to upgrade, provided there's enough memory to build the things.

Not that the wider business is struggling. Apple reported September-quarter revenue of $114.5 billion, up 12 percent year-on-year, while iPhone revenue climbed 17 percent to $57.6 billion. Services brought in $31.6 billion, up 10 percent.

Apple forecasts revenue growth of 7 to 10 percent in the December quarter despite the supply constraints. Even with memory prices running wild, it isn't expecting to go hungry. ®

Euro cloud trade association CISPE can keep the identities of members backing its VMware merger challenge confidential from Broadcom following a ruling by the EU General Court.

Cloud Infrastructure Services Providers in Europe filed an action last year seeking to annul the European Commission's approval of Broadcom's acquisition of the cloud and virtualization giant.

CISPE believes the Commission erred in allowing the purchase to go ahead without imposing any conditions on Broadcom to protect the many European cloud providers and their customers who depend on VMware's software.

Broadcom sought access to redacted information identifying members supporting the action, arguing that it needed the information to challenge the action's admissibility. CISPE says the ruling allows the case to proceed without disclosing those identities to Broadcom.

"It is a procedural ruling on confidentiality rather than a decision on the substance of the case, but it removes the immediate obstacle to the action moving forward," a CISPE spokesperson told The Register.

CISPE argued that identifying the members, which are or have been VMware customers, would expose them to a significant risk of retaliation because of Broadcom's market power. Those members had supplied information about their commercial relationships with VMware.

As far as the case itself goes, CISPE says it warned the Commission that Broadcom would use VMware's pre-eminent position in the private cloud market to rewrite licensing terms and reshape the market in its favor.

Just a month after Broadcom's acquisition concluded in November 2023, partners were informed that VMware's channel program was to end in early 2024, and that Broadcom was also ditching the majority of VMware's Cloud Services Providers, keeping only a select few that had the largest turnover.

Despite Broadcom CEO Hock Tan publicly stating that inflating prices was not his strategy, that is what happened. AT&T says that Broadcom proposed a 1,050 percent price increase, for example. Meanwhile, nine out of ten respondents to a recent survey of VMware customers said they were considering moving to alternatives.

CISPE, which represents 48 cloud infrastructure service providers across the EU, says fewer than 20 European providers remain authorized to offer VMware-based services, down from more than a thousand.

Commenting on the latest ruling, CISPE Secretary General Francisco Mingorance declared it a good day for the organization and for the rule of law in Europe.

"The case matters even more today than when we filed it. In the three years since the European Commission waved this deal through without conditions, Broadcom has dismantled the European VMware channel," he said.

"Those excluded have until 31 March, 2027, to sell their infrastructure or their customers to a competitor chosen by Broadcom, or to close. This is the foreclosure the European Commission was warned about."

We asked CISPE what outcome it hoped to achieve, nearly three years after the acquisition closed.

"The immediate objective is to have the Commission's unconditional clearance annulled, on the basis that it failed to properly assess the effect of the acquisition on European cloud providers and the VMware channel," CISPE's spokesperson told The Register.

The Commission did impose conditions concerning interoperability with rival hardware, but CISPE argues these did not address the risks to cloud providers.

"We appreciate that it's a long shot (although there is precedent)," the spokesperson continued. "We have (along with others) also filed a competition complaint with the Directorate-General for Competition against Broadcom for its actions post-merger."

The precedent referred to is the EC approval given to the merger of Sony and BMG's music divisions in 2004, which was ruled to be not of "the requisite legal standard" two years later. The deal was subsequently reassessed and re-cleared.

"Ultimately, we want to do whatever we can to bring Broadcom to the table to negotiate and ideally end some of its most egregious practices so that European cloud providers can once again support cloud customers in a commercially fair and viable manner," the spokesperson added.

The Register asked both the European Commission and Broadcom to comment, but neither had responded by the time of publication. ®

Public bodies in the UK expect higher operating costs and energy charges as a result of climate changes and the transition to lower carbon energy, but few have a good picture of the risks, according to government auditors.

A new report from the National Audit Office (NAO) says some public bodies expect financial impact from the risks they identified, although bodies were often not able to cost these - even approximately.

"Bodies anticipate costs to arise from damage to assets and infrastructure, driven by physical damage from extreme weather, as well as from chronic shifts to higher temperatures and more precipitation. Bodies also expect higher maintenance and repair costs, potential asset upgrades, and transition costs. Many bodies expect increased operating costs, including energy charges. Bodies controlling significant property or infrastructure are particularly concerned with decarbonization and adaptation costs," the report stated.

His Majesty's Treasury (HMT) has set out rules for climate-related financial disclosure in larger central government bodies' annual reports, which produced their first set of reports for the 2025-26 financial year. Other rules for disclosure have been in place since 2022.

However, the NAO found [PDF] that understanding of the risks in the impact of climate change - increasing temperatures, higher risk of flooding, higher winds and longer droughts - was poor, as was insight into costs of moving to a low-carbon economy.

"Bodies did not articulate transition risks as well as physical risks in the strategy disclosures. In the disclosures we reviewed, while several transition risks were listed, few specifics were given, unlike physical risks, which were explained in greater detail," the report from the watchdog stated.

"The transition risks most frequently identified included the financial cost of transition, and supply chain disruption. Anticipated transition costs included the costs of: switching to lower-carbon energy sources, including investments in new technology; complying with new regulations; and of fluctuations in the price and availability of materials, water, energy and specialist skills," it added.

Gareth Davies, head of the NAO, said: "UK public services will need to adapt to the impacts of climate change. Parliament needs clear information on the resilience of infrastructure, service provision and finances to climate change, and on how these risks are being managed."

"Although public bodies are still learning how best to report their most pressing climate risks, the requirement on all large central government bodies to make such disclosures from 2025-26 is an important first step. We expect this reporting to become more useful over time."

Earlier this year a report from Global Action Plan, a climate charity, claimed 84 percent of proposed datacenter developments were in areas that are already water stressed or projected to be water stressed by 2040.

Water consumption by datacenters could intensify strain on the UK's water systems already stressed by climate change, it argued.®

Raspberry Pi has announced yet another round of price hikes, this time adding $80 to the cost of its 16GB Pi 5, which now sells for more than three times its original launch price.

The latest bump, announced by Raspberry Pi boss Eben Upton, takes the 16GB Pi 5 to a whopping $385. The 16GB Compute Module 5 variants are also getting $40 more expensive, although the Raspberry Pi 500+, which packs the same amount of RAM, has escaped this particular round.

Upton blamed a familiar culprit: soaring memory costs, worsened by a lack of alternative suppliers for 16GB DRAM and ongoing shortages of other components.

"Unfortunately, a lack of alternative supply for 16GB DRAM, compounded by other ongoing component shortages, means we now need to make further price increases on some of our highest-density products," Upton wrote.

It's becoming an expensive habit for Raspberry Pi customers. The 16GB Pi 5 cost $120 when it launched in January 2025, which seems almost reasonable now. Raspberry Pi added $25 in December, another $60 in February, and a further $100 in April. With today's $80 increase, buyers are being asked to cough up $265 more than they were less than two years ago, a price hike of 221 percent.

And it's barely been a month since Raspberry Pi raised prices on its 2GB Pi 4 and Pi 5 models by $12.50 apiece, taking them to $67.50 and $77.50 respectively.

The repeated markups come despite Raspberry Pi reporting record first-half results in September. Revenue jumped 90 percent to $256.9 million, while gross profit rose 79 percent to $59.4 million. Gross profit per board climbed 53 percent to $12.20, helped partly by memory purchased before prices went through the roof. That stockpile advantage has since largely disappeared.

For customers unwilling to pay nearly $400 for a single-board computer, Upton advises buying something with less RAM, using an older Pi, or optimizing applications to make do with less memory.

He also warned that some sellers are exploiting the shortages by charging inflated prices or offering products that aren't what they appear to be, urging customers to stick to approved resellers.

"We're aware that a few bad actors have been taking advantage of the current memory supply circumstances to sell Raspberry Pi products at inflated prices or shift products that aren't what they seem," he said.

There is, apparently, some good news, as Raspberry Pi continues to insist the increases are temporary. "As we've said from the beginning, the current memory landscape is difficult but ultimately temporary, and we look forward to rolling back our price increases when memory costs fall once again," Upton wrote.

That could be a long wait. Micron has previously warned that memory shortages could drag on into 2028 or beyond, so anyone hoping to see a $120 Pi 5 again probably shouldn't hold their breath. ®

Distributed teams need to collaborate on shared information, but duplicated files, delays, and competing versions keep getting in the way.

Picture a video production team with editors in different cities. One person updates a project file, another keeps working from an earlier copy, and a third waits for a large transfer to finish.

When teams need to collaborate on large files across locations, traditional storage and file-sharing arrangements introduce friction precisely where people expect technology to help.

The hard part isn't merely moving a file. It's knowing who has the authoritative copy, when changes are available, and whether somebody else is still working on an outdated version.

Consider a game studio preparing a build for review. A designer exports an updated asset, but a colleague pulls an older version from a synced folder. The mistake only becomes obvious when the pieces are assembled. The team must trace the mismatch, resend files, and repeat work.

Sync-and-download workflows depend on a chain of uploads, downloads, and decisions about which copy is current. Under deadline pressure, that chain breaks. Delayed access means missed handoffs; duplicated files mean extra checking and sometimes rework.

For IT, each workaround creates another question: where does the data live, who can access it, and how do teams maintain consistent controls? Adding another collaboration application doesn't resolve those underlying questions.

The useful starting point is the workflow. Find out where teams copy data, where they wait, and what prompts them to work outside the shared system.

These frustrations cut across industries, but the ways organizations respond aren't identical. Hearing how another team handled version confusion, access, and handoffs is often more useful than another abstract promise of seamless collaboration.

That's the focus of The Register and LucidLink's invitation-only dinner on November 24 at Covent Garden Hotel in London. Joe Fay of The Register will moderate a peer discussion under the Chatham House Rule.

Bring the real situations that never fit neatly into a technology diagram, and hear how others are handling them.

Request your place at the November 24 dinner

Sponsored by LucidLink

The outputs of AI models are the result of a complex interplay between neurons, weights, training, and inference, and are therefore difficult to follow.

But in some ways they're starting to look more like open books. Recent work by Anthropic safety fellow Joshua Fonseca Rivera has revealed how certain neurons in a neural network can be associated with a specific persona.

Now, computer scientists from MIT's Computer Science and Artificial Intelligence Laboratory (CSAIL) have found that frontier AI models like OpenAI's GPT-6 Astra can rewrite the weights of another model directly, as opposed to indirectly through training, and alter their behavior.

"We find early signs that frontier LLMs are surprisingly good at natively reading, writing, and reasoning about neural network weights," said Rikiya Takehi, a PhD student at MIT CSAIL, in a social media post. "Instead of training, models may treat raw weights like readable, editable code: diagnosing failures and patching them directly."

To demonstrate that, Takehi and colleagues Jayden Teoh, Alex Zhang, and Omar Khattab asked GPT-6 Astra to edit Qwen3.5-0.8B weights to make it better at math.

Astra changed only 0.016 percent of the model's weights, reducing its reasoning loops and improving its math performance by as much as sixfold on the MATH-500 benchmark, while boosting its instruction-following score on the IFEval benchmark roughly threefold.

There was no training using gradient descent; Astra just revised the numerical values that constitute Qwen's weights.

"All of this is possible without allowing Astra to write training loops," the researchers wrote in a blog post. "It uses its reasoning to figure out what each part of a neural network computes, or should compute, and changes the weights accordingly. We call this Weight-Space Programming.

"Throughout the history of deep learning, we encoded our simplistic inductive biases into tiny symbolic optimizers like Adam or Muon. This is changing: our neural network weights will soon be natively optimized by LLMs that deliberately reason about their weights."

Astra's ability to program weights extends to being able to write a neural network from scratch, zero-shot. The researchers told Astra: "write, from scratch, the weights of a policy network that plays [Breakout] as well as possible." And it did so.

Takehi suggests this technique shows promise for improving models without the consequences of catastrophic forgetting, which can occur when a model is revised to include new knowledge.

More generally, the authors expect Weight-Space Programming will open the door to new ways to understand models and optimize them, and to implement recursive self-improvement. The technique may also prove useful for reducing model sycophancy and improving interpretability and alignment.

The next step, the authors say, involves building a benchmark and a harness for carrying out Weight-Space Programming on LLMs. "We believe this will provide a standardized evaluation to measure and accelerate progress in Weight-Space Programming, turning what currently seems like black magic into a rigorous, reproducible engineering discipline," they say. ®

For-profit model-makers, many of which are spending billions on AI infrastructure, will be chasing revenue from just ten percent of steps involved in agentic AI workflows, according to analyst firm Gartner.

"Nine in 10 agent workflow steps are routine, and open-weight models handle them at parity for a quarter of the price of the cheapest closed model," the firm's senior director Mark Petty and group vice president Raymond Paquet wrote last week in new document titled "Open-Weight Models Will Run 90% of Enterprise Agentic Workflow Steps."

"Typical operational workflows involve hundreds of routine executions, such as classification, tool calling, passing, validating and reconciling data," the pair observe, before suggesting open-weight models "can handle hundreds of these routine steps for every one that needs frontier reasoning."

"Against the cheapest closed model at the same score, the open model costs a quarter as much. Against a frontier model run on every step, it costs as little as a twentieth," the pair add.

Those are the best-case numbers, because the analysts admit "Current open models can be more verbose and retry more often, so the saving per task against a frontier model is closer to five or six times."

That's still a lovely saving, but the analysts rate it as the welcome by-product of moving to open models rather than an end in itself.

"Years of workflow history are a large part of what an acquirer pays for in a SaaS company and every routine step sent to a frontier API moves a slice of it outside your perimeter," the pair wrote. "Open-weights preserve your control. The cost saving is a bonus."

The two analysts therefore offer the following strategic planning assumption: "By 2030, 90% of enterprise operational agent workflow steps will execute on open-weight models."

They argue that enterprises will mostly adopt domain-specific open-weight models, because they do well "in scenarios where a small number of highly stable steps run millions of times a month."

Most models, however, are already commodities. Harnesses, the tools that run agents and manage agents, are the source of value in AI products, because one of their roles is to route work to the most appropriate model for a job, without users noticing any impact. A good harness saves money, imposes strong guardrails, and manages agent fleets. Less powerful harnesses leave users with more to do.

The paper suggests frontier models will continue to have a role as the engine of research and development and are effectively capital expense. Open-weight models become an operational expense after frontier models do the heavy lifting. ®

A cloud architect has created an open-source alternative to VMware's Virtual Disk Development Kit (VDDK), which the virtualization giant recently withdrew from public downloads.

VMware describes VDDK as "a set of libraries and utilities that enable software developers to create applications that can access and manipulate virtual disks" and "a valuable tool for developers of storage and backup applications."

Users, along with VMware's rivals, had another use for VDDK: migrating VMs to alternative virtualization platforms.

In August, VMware restricted access to VDDK downloads to its partners, meaning VMware customers couldn't access the tool and some of its rivals' migration tools became less effective.

Cloud architect Andrija Panić decided to do something about it.

"Many tools that have nothing in common except VMware use the same small library to read a virtual machine's disks: virt-v2v and nbdkit, the importers in Forklift/MTV, Proxmox VE and Apache CloudStack, and countless in-house scripts," he told The Register by email. "When public VDDK downloads ended, all of them became harder to install and run."

"I didn't think it was fair that people's ability to move or protect their own data should depend on a download page."

He describes the project as "an independent drop-in for libvixDiskLib.so.8.0.3, the disk library that VMware backup and migration tools load to read virtual disks. It was written from scratch, using only public documentation and black-box testing, and contains no VMware code."

Panić told The Register he wrote the tool in Go, with a thin C layer that exposes exactly the same C interface, library name and symbol versions as VMware's code.

The tool cannot write disks, just read them for backup and migration purposes.

In his email to The Register, Panić said he has successfully tested OpenVDDK on vSphere versions 6.7 U3, 7.0.3, 8.0.3, 9.1.1. He said that tests in his lab with the latter version, accessing OpenVDDK through vCenter on a standalone ESXi host, saw every disk read match.

"Independent testers have reported successful use with Forklift (VMware to KubeVirt, cold and warm), MTV on OpenShift (cold and warm), a Proxmox VE pipeline and custom nbdkit-based tools," the architect told The Register. "I've received reports of production use, including a German cloud provider migrating a large file server with V2K Migrate and a US-based company migrating to Apache CloudStack/KVM."

Panić was at pains to point out that the reports of successful use are "not independently audited customer endorsements or a guarantee of suitability for every production environment." As is often the case with FOSS tools.

OpenVDDK is not the only option for migrating away from VMware without VDDK. KVM-based OpenNebula has a tool called OneSwap that it says has a 96 percent success rate. Nutanix recently tweaked its "Move" migration tool so it can do the job without VDDK. Red Hat is working on a fix but has provided details of a workaround.

Panić made OpenVDDK available under an Apache license, so perhaps Red Hat and others will find it useful as they re-jig their VM migration tools. ®