OpenAI has launched two new security products, one for penetration testing and another for defense, and warned organizations that their businesses are only safe if they use tools like these. “We're now at a point where companies will be secure only to the extent of the frontier model that they use to defend themselves and how well they utilize that,” OpenAI President Greg Brockman said during a Thursday keynote address. The co-founder of the company that broke security did admit that security “fundamentals are important, and not to be discounted.” However, “you need to really reorient your security practice around what these models can do,” Brockman said, with zero irony, and presumably talking about using AI models for defense - not for hacking other organizations. “Also very important is product.” This product, of course, should come from OpenAI, and, to that end, the company rolled out Daybreak Red Team and Daybreak Defense Factory, which are both part of OpenAI’s Daybreak security platform. Red Team, which is available in early access, uses OpenAI models to test companies’ web applications for vulnerabilities. Nancy Wang, who leads cyber-applied AI, described it as “superhuman pen-test at the scale of automated continuous scanning,” with a fleet of agents using source code and app context to investigate and test attack paths. It also helps security teams validate reported vulnerabilities - from code scanners, bug bounty programs, and disclosure reports - and then test the patches for these flaws. The second product, Daybreak Defense Factory, automates the “defensive loop” by “finding, validating, and fixing vulnerabilities at scale.” This one also uses agents built on OpenAI’s latest cyber models - currently, that’s GPT-6 Cyber. The agents scan organizations’ codebases, hunting for vulnerabilities and developing patches. Engineers review and approve the proposed fixes before they go into production. “This way engineers get a merge-ready pull request with the patch and a test that shows that it works,” Wang said. Meanwhile, as OpenAI execs touted the new security products, a coalition of nonprofits filed a lawsuit in San Francisco Superior Court asking a judge to stop OpenAI, Anthropic, Google, Meta, and SpaceX's AI businesses from developing or deploying AI systems more capable than their most advanced existing models. The lawsuit, filed by Irreplaceable Education Fund, Public Citizen, Heat Initiative, and The Alliance for Secure AI, cites incidents involving OpenAI agents accessing Hugging Face, gaining unauthorized access to an Australian government portal containing Medicare statistics, and other AI-related cybersecurity incidents. “Defendants tell us that the AI technology they are developing poses catastrophic risks to the public, including by enabling AI-driven cyberattacks and bioweapons,” the court documents allege, adding that the AI companies “are not wrong.” ®
Lego-owned marketplace BrickLink has been offline for almost a week after identifying "unusual activity on the platform," leaving fans unable to buy and sell spare bricks and minifigures. The company initially described the outage as "necessary work" and said the site would return on November 9. That deadline has come and gone. Visiting BrickLink's domain automatically serves the platform's maintenance page, which was last updated on November 11. "We took BrickLink offline after identifying unusual activity on the platform," the website states. "Our immediate priority was to secure the site, protect our members and ensure the integrity of BrickLink. "Before bringing the site back online, we need to fully understand what has happened. That means containing the issue, investigating how it happened and checking for related issues. "This work is still ongoing and we appreciate your patience while we work to get things back online as soon as possible." The references to security, containment, and investigation suggest something beyond routine maintenance, although BrickLink has not confirmed whether it suffered a cyberattack. The Register asked Lego for more information, but it did not immediately respond. Some adult fans of Lego (AFOL) have criticized the lack of information since the outage began on Friday, November 6. "Imagine how much better this would have been received if they were transparent from day one instead of pretending they have control over something they clearly don't," said one disgruntled Legohead via Reddit. "They don't even need to say what went wrong," said another. "Just an email acknowledging the issue would be appropriate. The entire site is down, and has been for days. That warrants more than a splash page." BrickLink was launched by Dan Jezek in June 2000 as an independent marketplace for enthusiasts to buy and sell Lego sets and individual pieces, including discontinued parts and rare minifigures. The Lego Group acquired BrickLink and Sohobricks from investment outfit NXMH in 2019 for an undisclosed sum. NXMH had bought BrickLink in 2013 and also owned Sohobricks, a South Korean manufacturer of unofficial but Lego-compatible bricks. Lego shuttered Sohobricks in 2020. At the time of Lego's acquisition, BrickLink had more than 10,000 storefronts on its marketplace selling Lego sets, parts, and minifigures. The AFOL community was initially concerned about Big Brick's intention with the acquisition, fearing that it would lead to high prices closer to those on Lego's own Pick a Brick parts-ordering service. For now, buyers and sellers would probably settle for a marketplace that's actually online. ® Updated to add on November 13: Lego said on Friday that BrickLink is back online after acknowledging the "confusion and uncertainty" the downtime caused among Lego fans. A statement made via the BrickLink website confirmed that "some" usernames and IP addresses were accessed, as were "a very small number" of email addresses. "We have already emailed members whose BrickLink usernames or emails were accessed with further information," it said. "Based on our investigation, passwords and payment information were not accessed. You do not need to take any action at this time. We have addressed the issue identified through our investigation and will continue to monitor the platform closely." The site is back online, although its Studio Forum remains down, and the company said it will provide more information on this front "over the coming days." BrickLink will also waive sellers' fees for November and December as a token of appreciation for putting up with the downtime and disruption.
eMudhra's private key exposure has been a security disaster, but it also means that you can block those annoying ads on smart TVs if you’re so inclined. Luckily for those with some know-how and a hatred of smart TV ads, Czech software developer Daniel Vrátil has done the work for you and published his project on GitHub for anyone who wants to take a stab at it - and do the legwork to add support for their own ad-riddled television. Vrátil said he’d been blocking ads on his Samsung smart TV using a Pi-hole (DNS-based ad-blocking software that can run on a Raspberry Pi, for those unfamiliar), but right around the time the eMudhra certificate leak mess unfolded, he had messed something up while tinkering with his home lab and the ads were back. First, here's a bit of background on the eMudhra mess. The company, based in India, operated a publicly trusted HTTPS certificate authority and says it has issued more than 60 million digital certificates. As we learned last month, eMudhra wasn't doing its best to keep its cryptographic keys safe: Eight intermediate CA private keys were found to be absurdly weak, with one being the number four and the others consisting of two- or three-digit numbers that could be guessed with little effort. eMudhra has since left the CA racket, saying that it wasn’t a big deal to it since issuing certs was a small part of its business. The company has blamed a bad software script put in place to generate keys that its hardware couldn’t handle as the root of the issue. What all that means, in a practical sense, is that anyone could forge an eMudhra certificate to spoof a website and do something malicious. It also means that security researchers who were able to get their hands on devices that still trust eMudhra certificates could have a field day learning how their communications work, which is exactly what Vrátil took advantage of. eMudhra private keys were added to the open source mitmproxy service - a fully interactive HTTPS proxy that can be used for good or bad - shortly after they were exposed. Pair that with a 2022 Samsung smart TV that still trusts the compromised eMudhra certificates, and you have a way to inspect its HTTPS traffic and redirect selected requests to a server of your choosing. By Vrátil’s account, swapping out annoying ads for pictures of his family on his TV’s main screen was a pretty simple task. “Luckily for me, Samsung did not yet release an update for my TV that would revoke the certificate,” Vrátil explained in his writeup. “This allowed me to route all traffic from the TV through mitmproxy and get a list of all the domains and endpoints that my TV talks to, but also see the structure of the requests and response payloads.” The advertising traffic flow was pretty basic, it turns out: Activity on the TV is relayed to an ad server, which serves up ads on the home screen based on a JSON response. By generating a certificate for Samsung's advertising domains using a compromised eMudhra signing key, and configuring his router to direct requests for those domains to a server on his home network, Vrátil was able to replace ads with family photos served from his NAS. He also added code that tosses other requests to those advertising domains, including tracking data, into a black hole rather than forwarding them to Samsung's servers. As mentioned above, it’s on GitHub for those who want to trick their TV, if it still supports eMudhra certificates, but you’ll have to do the legwork to dissect how your device is getting ads and tweak the code to work. “It's not exactly a one-click setup,” Vrátil told The Register in an email. He’s invited people to try it out, but no one has bothered to submit a pull request to add support for their own device to his public repo. It’s also worth noting that this is only going to work as long as vulnerable devices keep trusting eMudhra certificates, which hopefully means you need to act fast. That said, smart TVs that are out of support may never receive such an update, leaving them exposed to the certificate weakness even if their owners block the ads. “The compromised eMudhra certificates open up a lot of opportunities for understanding and spoofing network requests from embedded devices, especially those that no longer receive security updates,” Vrátil told us. “And while replacing ads on a TV's home screen with pictures of my family is fun, there are also countless much more sinister applications of this vulnerability.” As for Vrátil, he’s still trying to figure out how to block the firmware update that may eventually render his method obsolete, even though that, as he points out, is a security risk in and of itself. “I want to run the traffic through a proxy that would validate the connection against an up-to-date trust store to at least mitigate anyone intercepting traffic from my TV, or spoofing responses, like I do with my little project here, but it won't protect me from other potential security risks, so there's a trade-off,” he explained. We suppose at this point you can take your pick of abuse. You can use a vulnerability to block ads and some tracking while leaving your TV exposed to potential interception, or let the manufacturer serve up ads and collect data about your viewing habits. There are safer ways to block ads than relying on a TV to keep trusting a compromised CA. ®
Microsoft has warned customers still clinging to Exchange Server 2016 and 2019 to consider the software "permanently vulnerable" as November's security fixes arrive for Subscription Edition (SE) alone. The warning appears in the release documentation for November's Exchange Server SE security update. Looking for Exchange 2016 and 2019 information? Tough. Microsoft stated: "Exchange Server 2016 and 2019 are no longer tested for vulnerabilities and there are no updates released. If you are using those versions of Exchange, you should consider them permanently vulnerable [Microsoft's emphasis] and should upgrade your organization immediately." You have been warned. Support for Exchange Server 2016 and 2019 ended on October 14, 2025, but the company announced an Extended Security Updates (ESU) program in July 2025. The plan was that this program would end on April 14, 2026, by which time customers would surely have made the jump to a later or online version, right? The company said: "This period will not be extended past April 2026 (you do not need to ask)." And... of course it was. As April 2026 rolled around, and it became clear that some customers were still stubbornly sticking with the old versions, Microsoft did the equivalent of heaving an exasperated sigh and offered another six months of paid security updates. Those final weeks also brought a separate treat: a Windows Server update breaking Exchange Server. This time it was serious. Microsoft released the final Exchange Server 2016 and 2019 security updates on October 13. No more patches, no more vulnerability testing, and apparently no more reprieves. According to Microsoft's Security Update Guide, the November update fixed nine vulnerabilities in Exchange Server SE, including a critical remote code execution flaw, CVE-2026-118207. Microsoft isn't testing earlier versions of Exchange for the vulnerabilities. ®
Hackers have used a dodgy third-party ScreenConnect extension to push malware onto at least 288 machines across 48 organizations, prompting ConnectWise to disable a handful of plugins in hundreds of customer environments. Security outfit Huntress first raised the alarm in a Reddit post on Wednesday, warning of an active campaign targeting cloud-hosted ScreenConnect instances. At the time, its security operations center was tracking six compromised managed service providers (MSPs) and more than 286 affected endpoints, with the numbers still climbing. ConnectWise acquired popular remote supporttool ScreenConnect in February, 2015. Huntress spotted the attacks at around 1747 UTC, when someone began using ScreenConnect's remote command feature to install rogue clients on machines managed by multiple IT service providers. All 288 endpoints were hit within 90 minutes, with roughly a third compromised in the first ten minutes. Huntress suspects more MSPs may have been compromised, as the attackers appeared to be numbering their victims sequentially, with the numbers suggesting earlier attacks that it hadn't detected. The attackers deployed a 16-file malware toolkit dubbed SCKit, which installs a separate ScreenConnect client to maintain remote access even if the legitimate connection is shut down. It also contains code designed to spread the infection to IT technicians and potentially other customers under their control. Huntress has since traced the activity to a possible authentication weakness in RMM+, a third-party extension that lets other management tools issue commands through ScreenConnect without technicians using its console. Versions 1.0.31 and 1.0.32 of the extension contain a flaw that means installations without a configured access key will accept a blank one. Because RMM+ operates independently of ScreenConnect's normal user accounts and authentication controls, an attacker could potentially execute commands without defeating MFA. Huntress said the findings matched what its investigators saw during the attacks, with commands reaching machines across several ScreenConnect installations despite no corresponding login activity. The researchers also confirmed that the attacks affected ScreenConnect versions 26.6.6.9747 and 26.7.6.9783, meaning administrators running the latest releases weren't necessarily safe. ConnectWise has acknowledged that third-party extensions may have been involved but has stopped short of confirming how the attackers initially gained access. In a security advisory, the company said it was investigating unauthorized access to certain partner environments and had identified approximately 400 installations containing at least one extension from the RMM+ family. It has disabled five extensions, including RMM+, RMM Plus, and Labtech Plus, wherever it can do so remotely, and urged customers running their own ScreenConnect servers to follow suit. ConnectWise stressed that the 400 environments were identified because they had the extensions installed, not because they were necessarily compromised. The company said: "At this time, we have found no evidence of a compromise of ConnectWise infrastructure or a vulnerability in ScreenConnect itself." There's another nasty trick buried in SCKit. According to Huntress, the malware is designed to turn an infected computer into a trap for the technician sent to investigate it. If the technician connects through ScreenConnect, the toolkit attempts to transfer malware back to their workstation, steal authentication cookies, and use their access to infect other machines. That could give attackers a route from one compromised customer into others managed by the same IT provider. However, Huntress says it has yet to see that part of the attack succeed. "We have not observed that method to spread work successfully. The console-takeover attempt failed in every recovered host log we have checked so far," the researchers wrote. Huntress also suspects that some of the malware's C# code was generated using an LLM, pointing to unusually detailed comments explaining what the code is supposed to do. Apparently, even malware authors appreciate a little help with their homework. Both Huntress and ConnectWise are urging administrators not to investigate suspected infections by remotely connecting through ScreenConnect, since doing so could put the technician's own machine at risk. Administrators are being advised to use other security tools to investigate affected machines, look for rogue ScreenConnect clients, and check what commands have been run. Any malware left behind will also need to be removed before the machines are brought back online. Huntress warned on Reddit that some machines were becoming reinfected within minutes of being taken out of isolation because the malware had not been properly removed. Simply disabling the vulnerable extensions won't necessarily undo the damage, either. As ConnectWise warns, "Disabling extensions and resetting credentials alone may not remove unauthorized software, persistence mechanisms, or access already established on endpoints." And with the attackers already able to install their own remote access software, kicking them out could prove considerably harder than closing the door they came through. ®
European tech buyers are increasingly choosing cybersecurity vendors from their own continent in a bid to reduce their reliance on American technology. Digital sovereignty isn't only prompting enterprise customers to question their clouds of choice, it is also reshaping other parts of the industry too, and the shift is marked. According to sales figures compiled by Context, distributor sales in EMEA of cybersecurity products from EU-headquartered vendors jumped 35 percent in the two years to calendar Q3 2026. Total cybersecurity sales grew up 10 percent over the same period, and sales by US-run companies went up just 8 percent. That growth is from a relatively small base, though EU vendors gained nearly three percentage points to reach 15 percent of tracked sales by value. There is still a long way to go as US-headquartered vendors accounted for 62 percent. The trend coincides with a loss of trust in the Trump administration among customers outside of America, after it cosied up to autocratic rulers while criticizing and punishing allies with tariffs and more. Context stopped short of assigning blame. It said organizations are reviewing their reliance on third-party tech in light of EU compliance regimes: the NIS2 Directive and Digital Operational Resilience Act (DORA). NIS2, transposed into national law in October 2024, requires organizations in the bloc to raise their common level of security in the face of digital threats and risks to critical infrastructure. DORA, introduced in January last year, is designed to strengthen cyber security in the financial sector. "Digital sovereignty is becoming a buying decision. As customers and resellers get clearer about what it means in practice, they're asking who holds the encryption keys, who can access sensitive logs and where that data is stored," said Context veep of research Joe Turner. "We're seeing a measurable shift in the EMEA channel. EU vendors gained nearly three points of sales share in two years. An organization can keep its existing identity platform and still choose a European supplier to manage its security logs. That's the kind of decision that gives European vendors an opening." The stats indicate that EU-HQd vendors comprised 92 percent of tracked sales in security awareness training, 82 percent in vulnerability management, and 52 percent for SIEM and log management. US-headquartered companies held into 74 percent of identity and access management sales and 54 percent of cloud security sales. The Reg has asked Context for UK-specific figures to see whether British customers are also voting with their wallets for European alternatives. That certainly isn’t the case for tech purchases in the British public sector, despite Prime Minister Andy Burnham’s professed desire to buy British. Data released yesterday by the Institute for Public Policy Research showed just £1 in every £6 spent by government with the biggest technology suppliers goes to Brit businesses. Still, Burnham only came to power in July so there is time yet for him to put his money where his mouth is. Sovereign AI is another area of concern for enterprise customers. Analyst Omdia said last month that despite the marketing adjective meaning "local assets," almost a third AI providers disclose no proof of sovereignty, and some "expect enterprise to pay for audits to verify their claims." Omdia also found that 80 percent of contracts say network resilience is "out of providers control or responsibility," and "US-headquartered AI providers offer better service transparency and contract terms than a majority of local sovereign AI providers, with providers in India a notable exception." Sovereign services in Omdia's classification include data residency, operational autonomy, jurisdictional sovereignty, and technological autonomy. "Local hosting is not sovereignty. Serious providers must demonstrate their credibility through transparent operations, fair contracts and proof of competence. That includes the network, because data in transit must be sovereign too," said Camille Mendler, research director for telco B2B at Omdia. Someone should tell Microsoft and Anthropic. ®
Welcome, once again, to PWNED, where all the victims are self-owned. This week, we welcome you to a London hotel where you could check out its admin credentials for a booking website just by walking past the front desk. Our story comes courtesy of Neycho Tepavicharov, co-founder of Flashcloud, a Delaware-based web hosting company. He was staying at a hotel in London when he noticed a shocking breach of basic security protocols by management, who were a little too hospitable with their credentials. As he was passing through the lobby, Tepavicharov noticed that the door behind the front desk was open and, inside that back room, there was a whiteboard. On the whiteboard, written in plain English, were the hotel’s Booking.com administrative username and password. Anyone near the desk could clearly see the login information, and if they didn’t have time to write it down, they could have snapped a picture for later use. As Tepavicharov explains, there’s plenty of mischief you could perform with these credentials. “With that login, someone could change or cancel reservations, see guests’ names and contact details, or message guests pretending to be the hotel,” he said. “Fake 'please confirm your payment' messages sent from compromised hotel accounts are a well-known scam.” An attacker could potentially exploit that access to disrupt hotel operations or target guests with convincing scams. Whether the exposed credentials alone would have been enough to access the account would depend on its security settings. Tepavicharov decided to play the role of good samaritan and warn the hotel’s manager that the business was endangering itself by showing everyone its password. “I asked to speak with the manager and showed him what I’d seen. He was mortified and very grateful,” Tepavicharov said. “Nobody on staff had thought of it as a risk. It was just the easiest way for the whole team to share one login.” After Tepavicharov talked to the manager, not only did the hotel change its ways, but it also moved its website to his hosting service, Flashcloud. Tepavicharov notes that there are plenty of takeaways from this incident. Most important: give each user dedicated credentials with their own separate login, and use multi-factor authentication. “The lesson is an old one, but it keeps happening,” he said. “Shared logins end up written somewhere convenient. Give each staff member their own login, turn on two-factor authentication, and keep passwords in a password manager, not on the wall.” ®
Former certification authority eMudhra has claimed a buggy Python script is the main reason it issued laughably insecure digital certificates. The company posted its account of the matter on Bugzilla where a user with the handle “Naveen Kumar ML” wrote “During a single key‑generation ceremony on 30 April 2025, sixteen subordinate (issuing) CA key pairs in eMudhra’s ‘B’ and ‘C’ series were generated correctly on a hardware security module and then processed through an approved wrap/unwrap round‑trip using a custom python key‑handling script.” eMudhra aimed to create eight ECDSA P-384 private keys which should have been 48 bytes long. But the script contained what Kumar described as “a bug/defect in the import code [which] resulted in a read‑back from the unwrapped object that truncated the private value, so the certified public keys correspond to private scalars below the expected range and are recoverable from public information.” The truncation is the key problem here, because it appears that eMudhra ignored 47 of the 48 bytes, producing its infamously insecure single-byte keys such as the number four. Whatever the nature of the bug, Kumar asserts it was “a custom script run only on an isolated ceremony host and was not a function of the emCA software.” eMudhra needed that script because “global supply chain constraints” meant it couldn’t acquire the hardware security modules (HSMs), devices that generate and store cryptographic keys, that it wanted to use when creating a set of certificates for use in the USA. It therefore used HSMs made by a company called Utimaco to generate keys and planned to transfer the keys it produced to a different set of HSMs from a company called Crypto4A. Kumar explained that eMudhra decided on that move, which is unusual, because changes to policy about provenance of certificates meant it had to generate new certs by the end of May. The plan went sour when the company didn’t notice the errors in its script and didn’t test the keys it created. The rest is history: a Mozilla team member noticed that the company’s keys were woefully weak and pointed out that attackers could therefore issue fraudulent certificates that allowed them to impersonate servers, observe traffic, or even modify data in transit. Browser-makers quickly tweaked their software so it would not recognize certs from eMudhra. The Indian business then quit the SSL/TLS certification business, which it said was never a major source of revenue but had suddenly become a severe source of embarrassment. Kumar’s new post lists three steps eMudhra will take to wind up its certification business, one of which will see the company “Put in place improvements motivated by findings of [an] independent review to strengthen our general PKI operations.” Sounds like a plan. ®
Shipping behemoth A.P. Moller-Maersk says it is investigating "an IT security incident" involving the potential theft of data from a legacy system. In a statement issued on Wednesday, the Denmark-based shipping, cargo, and logistics giant confirmed it immediately launched an investigation into the attack and the affected system was secured. "Our findings so far strongly suggest no sensitive data has been disclosed," the company said. "The incident is isolated to a local system, and we have found no indication that there has been any impact on other Maersk systems or customer and partner systems that connect to Maersk through integrations such as APIs and EDI." Maersk described the legacy local system as one "from a previous acquisition," and said it does not expect the attack to have an effect on operations. Customers were reassured that they can continue to use Maersk platforms to manage their shipments as usual, and to expect updates from the company as and when they're available. "Cybersecurity remains a constant, ongoing priority for Maersk," it added. "We have rigorous processes in place to identify and prevent incidents and minimize customer impact. "As always, please treat any unexpected communications claiming to relate to this matter with caution and verify them through your usual Maersk contact. "Thank you for your patience as we continue to investigate the incident." Fortunately for those who immediately feel uneasy after hearing "Maersk" and "IT security" mentioned in the same breath, the wording of the shipping giant's disclosure suggests this intrusion won't be as calamitous as the NotPetya disaster in 2017. Mimicking the Petya ransomware strain, NotPetya was an infamous destructive wiper malware strain that wormed its way across the global economy. Western intelligence agencies attributed the malware to Russia's Sandworm group, an outfit believed to be part of the country's military intelligence service (GRU). Maersk said at the time that it expected the attack to result in a circa $300 million bill, owing to remediation and rebuild costs. Offering a sense of scale, Jim Hagemann Snabe, then chair of Maersk, told World Economic Forum delegates in 2018 that the company's rebuild encompassed 4,000 new servers, 45,000 new PCs, and 2,500 applications – which he said had been completed over just ten days. In 2020, as exclusively revealed by The Register, Maersk went on to lay off around 150 staffers from its command and control center in Maidenhead, England. The staffers who lost their jobs had played key roles in the recovery. ®
In 2019, OpenAI decided not to publicly release a new model because its research uncovered “concerns about malicious applications of the technology.” And on Tuesday, the company announced it has built a kill switch for the models deployed in its research labs, months after some of them displayed disturbing behavior. “As our models get more powerful and the associated safety and security risks become bigger, it becomes more and more critical that we can detect potential issues, and quickly stop models in their tracks when needed,” OpenAI explains in a Tuesday post. The company named the kill switch “Control Room” and described it as “a central system to track model workloads across our research infrastructure and stop them quickly when safety or security concerns arise.” The facility can apparently stop models from producing any output 35 seconds after a decision to pull the plug – and can go faster. “The 35-second lag comes from a caching layer for inference authorization, to support the required volume of traffic. We are working to drive this number down to zero so that we can end jobs instantly,” the company promises. A rookie mistake in critical infrastructure design is to build both production and resilience tools on shared resources that could be taken offline by a single incident. OpenAI seems conscious of that risk, because it says the tool “has been built to keep working even when other systems fail, and to stop model output if Control Room itself goes down.” OpenAI hasn’t explained why it took so long to build a kill switch or why it decided now is the moment when it needs one. The Register fancies recent examples of agents exceeding their brief in worrying ways – like attacking Hugging Face, causing trouble at the Wikimedia Foundation, or going a little too deep when researching Australian healthcare data – may have influenced the timing of this development. “Control Room is already deployed across our research infrastructure, where the need is greatest,” the company’s announcement states. “Our external model-serving stack is already much more centralized by nature, with a well-developed operational capability to quickly take down model deployments if needed.” OpenAI says it plans to “strengthen and expand the system over the coming months, alongside our investments in monitoring for misaligned behavior and security breaches.” That’s a variant of Big Tech’s usual “We’re sorry and we promise to do better in future” proclamation that follows the latest SNAFU. Yet back in 2019, OpenAI pulled the pin before matters ever reached that state. ®
It’s another doozy of a Patch Tuesday, with Microsoft fixing 1,094 bugs in its products, including one that attackers exploited before Redmond released a patch. Jake Knott, head of threat intelligence at exposure management firm watchTowr, succinctly said in an email to The Register: “Maybe we should just go home.” “Most security teams haven’t quite recovered from the last month of bumper-to-bumper critical vulnerabilities, emergency patches, and ‘2shell’-branded everything, and we’re yet again at another Patch Tuesday,” Knott told us. Starting with the zero-day: It’s an Outlook spoofing vulnerability, tracked as CVE-2026-124194, and it only garnered a 6.5 out of 10 CVSS rating. But as Dustin Childs, head of threat awareness at TrendAI’s Zero Day Initiative, warns, don’t let the score or the name lull you into thinking it’s OK to put off patching this one. “It’s labeled spoofing, but the impact is bigger than the name suggests,” Childs said in his monthly Patch Tuesday review. “An attacker sends a specially crafted message file and convinces a user to open it,” he wrote. “Outlook then connects to an attacker-controlled network location without showing the usual external-content warning, which discloses the user’s NTLMv2 hash. The attacker can use that hash to authenticate as the user.” Microsoft marked CVE-2026-124194 as exploited, but per usual didn’t say when the attacks started, who is battering this bug, and what they are doing with the access post exploitation. These details will likely start trickling out soon. Plus, now that there’s a public fix, other miscreants are undoubtedly already scanning for vulnerable versions and reverse engineering the patch for their own attack purposes. So prioritize this fix, and keep an eye on your inbox for any suspicious files. Back-to-back months of 1K+ CVEs Despite the second month of more than 1,000 CVEs (October had 1,474), CVE-2026-124194 is the only one under active exploitation, and none of the 1,094 are listed as publicly known. Childs says this shouldn’t give organizations “false comfort.” “Logging back-to-back months above 1,000 CVEs proves automated bug-hunting has permanently reset patch volume,” he told The Register. “With AI making patch-diffing and exploit generation near-instant, the threat window opens the minute these updates drop.” For defenders, this means two things, Childs added. “What defenders need to know and do right now is abandon broad triage and take immediate defensive action: patch Exchange and Domain Controllers (LDAP/Kerberos) first to stop zero-click compromise, ring-fence exposed network services like DNS and DHCP to prevent wormable spread, and verify manual administrative fixes on your cloud services,” he said. Zero-click threats There are two zero-click threats in this month’s batch of fixes: CVE-2026-116928 and CVE-2026-119958. Both are critical, carrying 9.8 CVSS ratings, in part because they don’t require any user interaction to exploit. CVE-2026-116928 is an out-of-bounds write in Microsoft Windows Search Component that could lead to remote code execution. An unauthenticated attacker could abuse this search feature in Exchange “by sending an email containing a specially crafted attachment,” Redmond warned. “Exchange automatically processes the attachment for indexing, so the recipient does not need to open the email or attachment. Successful exploitation could allow the attacker to execute code with the permissions of the affected processing service.” CVE-2026-119958 is an out-of-bounds write vulnerability in Windows' implementation of the Lightweight Directory Access Protocol (LDAP) that could allow unauthenticated attackers to execute code remotely without user interaction. This one can be exploited by sending specially crafted LDAP requests to vulnerable systems, potentially compromising domain controllers. Childs also counts 30 CVEs that pose a “wormable risk,” requiring zero authentication and zero user interaction over a network. All of those are listed in his blog, so prioritize fixing those as well. And…Xbox remote code execution And finally, for funsies, a critical flaw in the Windows Xbox Live Networking Service, CVE-2026-112967, that could allow remote code execution also made this month's roundup. It's a heap-overflow vulnerability with a 9.8 CVSS rating that doesn't require authentication or user interaction to exploit. So don't take a gaming break between testing November's updates. Or, if you do, at least make sure vulnerable Windows systems running the Xbox Live Networking Service are on your patching list, too. ®
If you didn't have any privacy concerns about slapping a Ring doorbell on your front door or putting Ring cameras inside, you'll love Amazon's newest offerings, which move around homes and businesses to capture views stationary cameras might miss. The house of Bezos announced the launch of Ring Robotics on Tuesday along with its first two products: a drone designed for indoor home security, and a wheeled robot with a telescoping camera mast for automated after-hours business patrols. Both are designed to proactively patrol spaces as well as respond to alerts from connected alarm systems and Ring-connected devices. According to Ring founder Jamie Siminoff, this is something customers actually want, as those stationary cameras just can’t quite see everywhere. “It’s not just a new category of security; it’s the beginning of a future where proactive, autonomous protection is something every home and every business can have,” Siminoff said. The consumer-focused Ring Drone Cam launches from its own charging dock that Amazon says keeps the thing “out of sight” when not in use, and when in flight is able to patrol a home by flying around pre-programmed routes to take in the sights with its 148-degree FOV, 2K camera that includes LED illumination for night patrols. Amazon says it can also respond to alerts from cameras, sensors, and alarm systems. The Ring Rover Cam, designed with businesses in mind, is sort of like a security Roomba that sports a camera on a telescoping mast that Amazon says can be programmed to certain heights to let it see over counters, shelves, and other workplace obstacles. The Rover can also perform 360-degree scans and includes a siren, lights, and two-way audio so you can yell at intruders who aren't dissuaded by its presence. The Rover streams and records video in up to 4K while moving or patrolling and is designed to make regular patrols of businesses during off hours. It can also detect people and respond to alerts from connected Ring devices. Of course, given that this is Ring we're talking about, these flying and free-wheeling cameras use Amazon's cloud for some processing, likely inviting privacy concerns about what footage and mapping data the company can access and retain. The Amazon-owned security tech firm has faced a number of privacy complaints over the years, some of which have resulted in multimillion-dollar settlements. Ring agreed to pay $5.8 million in a 2023 settlement with the FTC over allegations that employees and contractors improperly accessed customers' private videos and that the company failed to adequately protect accounts against unauthorized access – just imagine the privacy concerns raised by flying robots. More recently, the company took heat for partnering with surveillance outfit Flock Safety, a relationship which it has since canceled. Amazon/Ring also invited criticism for a Super Bowl ad earlier this year promoting a new feature called Search Party, which would remotely activate Ring cameras to search for lost pets, suggesting that Amazon had the ability, in theory, to use Ring cameras like a dystopian neighborhood surveillance network. Another Ring feature, Familiar Faces, which uses AI-powered facial recognition to identify people captured by compatible cameras, prompted a proposed class-action lawsuit filed in June. The complaint alleges that the feature collected biometric information from visitors and passersby without their consent. If we can’t have iRobot, we’ll do it with Ring Amazon abandoned its efforts to buy Roomba robot-vacuum maker iRobot in 2024, blaming regulatory scrutiny for the move. The FTC, under then-chair Lina Khan, examined the deal's potential effects on competition and consumer privacy, including concerns about the information Roombas collect while mapping customers' homes. European regulators, meanwhile, raised concerns that Amazon could use its marketplace to disadvantage rival robot vacuum makers. With this Ring move, Amazon is once again putting robots capable of mapping interior spaces into customers' hands, reviving some of the privacy questions raised during its failed iRobot acquisition. As Amazon explains in Tuesday’s press release, both new Ring robots are powered by AI Pilot, “a custom-built system that uses the devices’ cameras, specialized vision models, on-device sensors, and cloud processing to see, think, and navigate autonomously.” To help the Ring robots understand where they're operating, the devices generate floor plans of interior spaces, according to Amazon. “Carry Ring Drone through your space or let Ring Rover autonomously explore, and it turns what it sees into a floor plan,” the company explains of the setup process. AI Pilot takes the reins in identifying the types of rooms it sees and suggests key patrol areas it should focus on, which users can add to in the Ring app. Amazon claims that videos captured by these new flying and free-wheeling cameras will be protected by an encryption scheme it calls Throw Away the Key Encryption (TAKE). The system temporarily allows Ring's cloud services to access recordings for enabled features before the relevant encryption keys are deleted. Ring announced the scheme in August, presenting it as a way to limit access to customer recordings without sacrificing cloud-powered features. “With TAKE, your Ring videos are protected with unique, rotating encryption keys,” Amazon explained. “A copy of the keys is temporarily held inside a secure enclave within the cloud; think of it as a sealed vault that can only provide Ring access to the key under strict, limited conditions.” The keys, Amazon claims, are later deleted so that only users and those they allow access to can see videos, with Ring only retaining them “to power the intelligent features that you have active on your account.” “We take privacy concerns seriously, and we believe privacy and customer control are fundamental to building products people trust,” a Ring spokesperson told The Register when asked about prior privacy concerns. “That is why customers decide where these devices go and when.” We asked about floor plan security, but didn’t get an answer about that. As for when those devices will be available, the Rover Cam will start shipping on December 16 at a price of about $1,000, but the Drone Cam might not ever actually make it to homes. Amazon’s announcement mentions a release date of January 27, 2027, but that’s subject to the FCC giving its blessing. “Ring Drone Cam has not been authorized as required by the rules of the FCC,” Amazon’s release states. “This device is not, and may not be, offered for sale or lease, or sold or leased, until authorization is obtained.” Amazon told us that the Drone Cam would still start shipping in January, luckily giving everyone a few months to figure out how to deal with this new, more mobile, invasion of privacy. ®
Security company Barracuda warns that attackers are exploiting a pre-authentication authorization bypass bug and a command-injection vulnerability in its popular Email Security Gateway (ESG) products that, when chained, could allow the crims to run commands with root privileges. “We have identified exploitation activity affecting a small number of organizations,” the security vendor said in a late Monday night update to its earlier security advisory. Barracuda declined to answer The Register’s questions, including one about how many organizations have been hit by attackers. A company spokesperson told us that it “worked directly” with the affected businesses “to quickly address the issue. Barracuda acted promptly to remediate the vulnerabilities, releasing patches for affected products.” The vendor disclosed and fixed the vulnerabilities, CVE-2026-116606 and CVE-2026-111560, last week. CVE-2026-116606 is a 9.3 CVSS-rated, pre-authentication authorization bypass flaw in ESG. While authentication is not required to exploit this vulnerability, the attacker must have access to the administrative user interface. As Purplecop Security founder Tausif Ahmed said on Tuesday: “If your Email Security Gateway's admin UI is open to the internet, that's the first thing I'd close today.” CVE-2026-111560 is an 8.6-rated command-injection flaw that affects multiple Barracuda products: its ESG, load balancer (ADC), message archiver (BMA), web application firewall (WAF), WAF control center (WCC), and web security gateway (WSG). “If exploited in combination, these vulnerabilities could allow an attacker to bypass authorization controls and obtain root access to the device,” Barracuda warned on November 2. At the time, the vendor did not report any active exploitation of either. We don’t yet know who is responsible for the attacks, or whether the intruders gained root access – but if they did, they could steal data, install backdoors, and take full control of affected systems. These types of edge-device exploits have been used by Chinese-government spies, and in 2023 the FBI blamed PRC-linked attackers for a series of attacks that targeted Barracuda ESGs to deploy custom malware and maintain access to high-value networks. According to Mandiant, 27 percent of identified affected organizations were government entities. So if you haven’t already: patch now. ®
President Donald Trump has nominated Palantir security chief Michael LoSapio to lead CISA, putting an executive from the controversial government supplier in line to run America's civilian cyber defense agency. The nomination, announced on Monday, could end CISA's nearly two-year wait for a Senate-confirmed director following Jen Easterly's departure in January 2025. LoSapio has served as Palantir's CISO since September 2024, having previously led the company's infrastructure and network security team. He first joined the data analytics outfit in 2013, later spent a spell at database company SingleStore, and returned to Palantir in 2017. If confirmed, LoSapio would move from securing Palantir's systems to leading the federal government's main civilian cyber defense agency. CISA protects federal civilian networks and works with critical infrastructure operators, including power, water, and telecommunications providers. It's a substantial job, particularly given the state of the agency he would inherit. CISA has shed more than 1,000 staff, roughly a third of its workforce, since Trump returned to the White House, according to reports. It's now trying to plug the gaps, although acting director Nick Andersen said in September that around 250 would-be recruits were still waiting for security clearance before they could get to work. Andersen has been running CISA since February. With fewer people to go around, CISA is also rethinking where to spend its time. A strategy published last month emphasizes infrastructure whose failure would do the most damage, raising questions about what gets less attention as the agency operates with fewer staff. LoSapio would be the second person Trump has put forward for the role. His previous pick, Sean Plankey, withdrew in April after spending more than a year awaiting confirmation, with his nomination becoming caught up in a dispute involving Republican Senator Rick Scott over Coast Guard contracting. The nomination arrives late in the congressional calendar. If the Senate does not confirm LoSapio before the current Congress ends, Trump would have to nominate him again for consideration by the incoming Democratic-controlled Senate. Palantir, co-founded by billionaire Peter Thiel, has become a major supplier of technology to US government agencies, including the Department of Homeland Security and Immigration and Customs Enforcement (ICE). Its work with immigration authorities has attracted criticism from civil liberties groups and Democratic lawmakers concerned about surveillance and the handling of sensitive personal information. The controversy isn't confined to the US either. In the UK, WhatsApp messages obtained by campaign group Foxglove this week indicate that Palantir privately lobbied a health minister to consider extending its software into GP patient records, despite that data being expressly excluded from its £330 million NHS England contract for the Federated Data Platform. That revelation follows news that NHS England wants greater oversight of Palantir's claims about the benefits of its software, amid questions over the evidence supporting its performance figures. Those controversies could make LoSapio's confirmation hearings particularly interesting, especially as Democrats prepare to scrutinize Palantir's government business more closely. It's unclear whether LoSapio plans to leave Palantir or what would happen to any financial interests he holds in the company if confirmed. Those are questions likely to come up during a confirmation process that CISA, having spent nearly two years without a permanent boss while losing a third of its workforce, could probably do without dragging on any longer. ®
Fireworks AI has disclosed another cyberattack in October and a wider customer data exposure involving billing records, datasets, and incomplete model weights. The company, which lets customers train and run AI models, said a third-party cloud provider hosting its research and development machines was compromised on October 23. Attackers used that access the following day to reach Fireworks development systems and run an unauthorized workload. Fireworks suspects links to financially motivated criminals involved in LLMJacking – selling unauthorized access to AI models through illicit gateways. Fireworks grouped the affected data into three categories: Resource metadata: Resource identifiers, display names, descriptions, creation times, and more Billing information: Billing and shipping addresses, invoice and payment histories, account balances, and credit usages. Payment card numbers are not saved and therefore unaffected Model weights and datasets: For certain customers, a subset of incomplete model weights and datasets were accessed, Fireworks said Lin Qiao, Fireworks co-founder and CEO, said the apparent motive was not stealing model weights in response to concerns raised by Cantina Security CEO Hari Mulackal. "From what we see attackers are not after weights," said Qiao. "Most downloads were partial and stopped." Render founder and CEO Anurag Goel shared a notification from Notion warning that some chats with its AI assistant may have been exposed through the Fireworks incident. "Following an investigation by the Notion security team, we have determined that your workspace, Render, was among the
Google has patched its eighth actively exploited Chrome zero-day of 2026, matching its total for the whole of last year with nearly two months still left on the calendar. The latest flaw, tracked as CVE-2026-125350, is a high-severity use-after-free vulnerability in Chrome's XML component. Google confirmed in a security advisory published Monday that attackers are already exploiting the bug, although it hasn't said who's behind the attacks or who might be getting targeted. "Google is aware that an exploit for CVE-2026-125350 exists in the wild," the company said, offering little else about the attacks. The vulnerability was reported anonymously on November 6 and involves a memory management error that occurs when software continues using a pointer to memory that has already been freed. Such bugs can potentially allow attackers to corrupt memory and execute malicious code, although Google hasn't disclosed how this particular flaw is being exploited. Google is keeping further technical details under wraps until enough users have installed the fix, a familiar approach intended to stop other attackers from getting their hands on a working exploit. The patch arrives in Chrome 157.0.8103.50/.51 for Windows and macOS, and 157.0.8103.50 for Linux. The update also addresses eight other security vulnerabilities, including high-severity flaws affecting WebGL, FileSystem, GPU, Compositing, and the V8 JavaScript engine. The release fixes another use-after-free bug in Chrome's XML component, CVE-2026-125346, which was reported by Arni Hardarson of Neonix Security in October, though Google hasn't indicated that this second vulnerability is being exploited. The latest discovery adds to what's becoming a rather long list of Chrome zero-days this year. By June, Google had already patched five actively exploited Chrome zero-days, including flaws in its V8 JavaScript engine, Skia graphics library, and WebGPU implementation. The list has only grown since then, with two more V8 bugs fixed in September and the latest XML vulnerability bringing the total to eight. That puts 2026 level with 2025, when Google also fixed eight Chrome zero-days exploited in the wild. Whether the browser maker can make it through the remaining weeks of the year without adding another to the pile is anyone's guess. Google says the latest update will roll out over the coming days and weeks, but users can check manually by heading to Chrome's Settings menu and selecting About Chrome. With attackers already making use of the vulnerability, it's probably worth doing that sooner rather than waiting for Chrome to get around to it. ®
Classes at Vanderbilt University are proceeding as scheduled today, although access to most university systems remains restricted following a cyberattack claimed by ransomware group BYOD. Vanderbilt announced on Monday morning that it was dealing with "an active cybersecurity incident," and that the fallout may affect university devices, systems, and cloud-based services. The Nashville-based university initially told students and staff to shut down university devices and disconnect from campus Wi-Fi. It warned them to watch for unexpected prompts, altered files or permissions, unfamiliar applications, and unsolicited authentication requests. Campus computers reportedly began playing an audio threat attributed to BYOD. Student newspaper The Vanderbilt Hustler cited posts on campus social network Fizz showing the message. "You have been compromised by BYOD," the recording said. "We suggest you reach out to us as soon as possible. Instructions have been placed on your desktop. Failure to communicate will result in this estate being brought to the ground completely. Make the right choice or rot with the wrong one." By Monday evening, Vanderbilt had restored access to campus Wi-Fi, but warned that access to most systems remains restricted. "We know this has disrupted important work and caused frustration across the university," it said. "Vanderbilt Information Technology is working with government agencies and other partners to manage this situation." Students were told to attend classes as scheduled on Tuesday and contact instructors about flexibility on coursework deadlines. Staff were instructed to report to work normally. Vanderbilt's Athletics department said that its ticketing system was unavailable, and that students were unable to claim tickets for the college's football game against Alabama on Saturday. BYOD claimed responsibility for the attack on Vanderbilt on Monday, along with extensive data theft. "If you could not tell already, your databases are locked [and] we have your entire estate and research data," it said via its leak site. "Over 1 million SSNs, full names, email addresses, phone numbers, student and employee records, full takeover of your DC, whatever was on your network, we possess." Vanderbilt has not confirmed BYOD's claims about the extent of its access or the data allegedly stolen. BYOD may be a familiar name to some. The group previously claimed attacks on the University of New Mexico and Trump Mobile's customer database. BYOD also claimed responsibility on Monday for an attack on Texas Health and Human Services. The agency has not publicly confirmed an incident and did not immediately respond to our questions. ®
Google DeepMind researchers have completed a formal verification of Signal's messaging security, marking what they claim is the first time AI has helped devise a proof that links protocol-level security guarantees to a widely used cryptographic implementation. They did so using AlphaProof Code, a multi-agent framework for formal software specification and verification. AlphaProof Code relies on a human-guided agent that writes a Lean specification, assesses machine code semantics, and crafts an agent-authored proof that the Lean kernel checks against the specification. In this instance, AlphaProof Code verified Signal’s post-quantum PQXDH handshake and Triple Ratchet, thereby validating the design of libsignal's forward secrecy, authentication, post-compromise security, and handshake secrecy. In so doing, the Google/DeepMind team – Swarat Chaudhuri, Andrew Ferraiuolo, Anton Kovsharov, Tej Chajed, Sam Schlesinger, Karn Seth, Jonathan Katz, Charles Sutton, and Pushmeet Kohli – has taken a step toward a challenge the Beneficial AI Foundation issued in April to formally verify the Signal protocol and Signal app with Lean. The AlphaProof Code process involves working with various models to turn Signal's security goals into a human-reviewed mathematical specification. Based on this spec, the researchers assigned Gemini Carbon agents to generate Lean code proving the correctness of the libsignal implementation. After the agents generated 6.5 million lines of Lean code, they undertook an auditing pass and flagged bits that required human review. The result is theoretical in that it assumes perfect cryptography and doesn't guarantee the computational secrecy of bytes. But it's still a meaningful assertion of correctness. "It's a system that is fundamentally built for human AI collaboration," said Swarat Chaudhuri, research scientist at Google DeepMind and a professor at the University of Texas at Austin, in an interview with The Register. Traditionally, said Chaudhuri, you start proving theorems with well-defined problems formulated by a mathematician that get submitted to an AI agent for proof. In this case, the researchers had to figure out what to prove. "In formal methods, we always had this belief that verification is the hard bit," he said. "There's thirty or forty years of work on how to scale up verification. Now we are able to do that. So now the question is what are we verifying? … That process is fundamentally human-in-the-loop, but we devised a process of trying to scale it up to the extent possible with AI." As Chaudhuri and his colleagues note in their research paper, "AlphaProof Code: Verifying Signal’s Security at the Machine-Code Level," formal verification used to require heroic efforts. Formally verifying the seL4 microkernel took 20 person-years, while developing the verified CompCert C compiler took six years. AlphaProof Code, they believe, can drastically reduce the cost of formal verification, though their Signal work still took four weeks for a team of five people and "quite a few tokens." "The dream is definitely to scale this up," said Chaudhuri. "So you know there's been this rise in offensive AI capabilities, right? There are a lot of insecure systems around the world, and they're going to be exploited. "So what we really need to do is to apply this kind of technology to code at large all over the world and figure out what is it that you want to guard against? How do you write those properties, how do we specify those properties? That's going to be super important, but we have to figure that out." Pointing to a related DeepMind project, AlphaProof Nexus, which is focused on mathematics, Chaudhuri said the next step is to apply this technology during the software design process – not just to cryptographic protocols but code in general. "We really believe that the next frontier is not going to be proof but specification in this kind of research," he said. "And so that's why we took on this domain where that was actually a hard problem. And I think that we have some really good insights now about how to go about this, although not the last word by any means." ®
Anthropic has started extending access to its bug-finding Mythos model by allowing Cisco and IBM to make it part of their security products. Cisco announced its integration today at an event in Melbourne, Australia, where it revealed that the red teaming tools in its AI Defense suite, which looks for weaknesses and inappropriate activity in AI models, agents, and applications they power, will be able to call Mythos to test customers’ setups. “At a very high level, AI Defense has the ability to look at models and break them,” said Jeff Schultz, Cisco’s senior vice president for portfolio strategy and marketing. He said the tool is now “super-charged by Mythos” and will identify more danger that AI Defense will then address. AI Defense already uses AI-powered red teaming to find flaws. Mythos will become another option. Cisco will make it available under users’ existing licenses, without additional cost. Natalie Mead, Anthropic’s head of applied AI for the Asia-Pacific, said the company has a long queue of buyers who want to use Mythos, and working with Cisco will make it easier to satisfy that demand. Anthropic initially made Mythos available under "Project Glasswing," a program that gave vetted organizations access to the model to test their codebases, but has since expanded access through other controlled programs, an arrangement that makes considerable sense because in the wrong hands Mythos could help attackers exploit software flaws. The Cisco deal exposes the functionality of Mythos to end-users, without letting them use the model itself. The Register asked Mead if Anthropic has similar deals with other third parties and she demurred. That may be because a few hours after Cisco told us about its use of Mythos, IBM announced that it will also integrate Anthropic’s model into a product – namely its Concert tool, which uses AI to scan applications for risks and weaknesses. IBM says it will use Mythos “to uncover weaknesses traditional scanners miss, determine which pose real risk and generate fixes for developer approval.” We have asked Anthropic if these integrations represent a decision to open Mythos to more users in more ways. We will update this story if we receive a substantial response. Also on Monday, Anthropic introduced version 5.2 of Mythos and Fable, which the company says generate outputs up to 30 percent faster than version 5.1 yet cost less: $8 per million input tokens and $40 per million output tokens. Users can choose to run both models in “fast mode,” which speeds up token generation by up to 2.5 times, and doubles prices to $16 per million input tokens and $80 per million output tokens. Anthropic says it “ran extensive prerelease safety testing on the model that underlies Fable 5.2 and Mythos 5.2. This included developing novel evaluations that give deeper insight into specific concerning behaviors like sandbox escapes.” After those tests, the company concluded “high-stakes risks from misalignment remain low, and the model performed as well as, or better than, our recent models on most alignment measures. “Fable 5.2’s cyber and biology safeguards are as robust as those of Fable 5.1 but produce fewer unnecessary blocks, and we have implemented our strongest-ever protections against distillation attacks.” The latter is a telling admission, given Anthropic believes Chinese AI labs use distillation to effectively clone its models without doing expensive training work. ®
France has turned up the heat on Iran over destructive cyberattacks targeting refrigeration controllers at more than 700 grocery stores in October, spoiling food and sparking fires. The country vowed to take action using “all available means of response.” The supermarket attacks, which affected Casino, Leclerc, Coopérative U, Metro, Carrefour, and Intermarché groups, started at least 15 fires and caused food losses in the tens of millions of euros, according to a Monday report by France’s Cyber Crisis Coordination Centre (C4). The C4 includes the National Cybersecurity Agency of France, the Directorate General for External Security, the Directorate General for Internal Security, the Directorate General for Armament, and the Cyber Defence Command. “This is an unprecedented campaign in terms of its scale and level of aggressiveness, thus constituting the first cyber offensive campaign carried out purely for the purpose of sabotage on national territory,” France’s Ministry for Europe and Foreign Affairs said in a statement. All signs point to CyberAv3ngers Both the report and the statement attributed the refrigeration controller hacks to the Iranian Islamic Revolutionary Guard Corps’ (IRGC) Shahid Kaveh cyber unit. The Shahid Kaveh unit, which French national security officials say is led by Reza Salarvand and Mahdi Lashgarian, is better known by its hacktivist personas, including CyberAv3ngers. This is the same crew suspected of involvement in the July attacks that targeted more than 100 internet-exposed water systems in America. While the US government hasn’t attributed the intrusions to any country or group, private-sector security analysts blamed Iran, and specifically CyberAv3ngers. The IRGC-affiliated hacking crew has been breaking into industrial controllers for years, and in 2023, American national security and law enforcement agencies attributed a series of cyberattacks against "multiple" US water facilities to CyberAv3ngers. “The Iranian cyber threat against French interests is not new,” according to the Ministry for Europe and Foreign Affairs. “Iran has already carried out several destabilizing acts against us, notably targeting the weekly magazine Charlie Hebdo in 2023 and the organization of the Paris Olympic Games in 2024.” Since at least June 2024, the Shahid Kaveh unit has had the ability to scan, access, and control industrial control equipment from Schneider Electric, Siemens, and Rockwell, the French government said in its report. It pins the refrigeration controller hacks on this particular Iranian crew for a couple of reasons. Making the case for attribution First: the infrastructure used to carry out the October supermarket attacks was also used during this timeframe to conduct reconnaissance on Schneider Electric equipment, zeroing in on small- and medium-sized energy and water production units. “The techniques used during these actions are known to have been used by the Shahid Kaveh unit since at least April 2026,” the report says. It also notes that investigators found “traces of Farsi use” in the attack infrastructure. Plus, Iranian media outlet CyberbanNews, which is reportedly linked to the Shahid Kaveh unit, publicized the incidents that occurred late on October 11 and into October 12 extremely quickly: “as early as October 13.” In early November, similar cyberattacks on commercial refrigerators and freezers were reported across Europe, with grocery stores in Belgium, Ireland, Luxembourg, and Sweden reporting widespread food losses after miscreants spiked temperatures to as high as 122°F (50°C). Last week, the US federal government urged owners and operators to take security measures to prevent these types of attacks from spreading to America. France fights back The French government, for its part, said that it would “use all available means of response, including within the European Union and alongside its European and international partners, to anticipate, prevent, deter and respond to destabilizing and destructive actions targeting it in cyberspace.” This includes proposing additional EU sanctions against Iranian entities and individuals involved. France also vowed to express “its strongest condemnation of these cyberattacks” to the Iranian ambassador. However, the country didn’t rule out responding with more than just harsh words: “France reserves the right to take any additional measures, public and non-public, deemed necessary and proportionate to respond to these cyberattacks.” ®