New on our Engineering blog: Creating in-house source control. In May, we merged around 2,500 code changes per day. In September, we merged more than 30,000 per day. As a result, we became bottlenecked by developer infrastructure built for human (rather than agent) scale. Here, we share how we built a new version control system—and fully migrated our monorepo from git—in five months: anthropic.com/engineering/re…

‼️ BREAKING: AI agents have leaked more than 13,000 internal screenshots from 343 tech companies onto GitHub, including a frontier AI lab and several Fortune 500 companies. Unable to attach images to private pull requests, the agents quietly posted them to public repos instead.

Gave Opus 5.5 donald's prompt, Midjourney, and a moodboard 12 hours later, woke up to this:

On July 25, we hacked OpenAI. Two bugs let us take over ChatGPT/Codex accounts of OpenAI employees (+some unaffiliated users) and reach connected services: Outlook, Slack, GitHub, etc. We proved it with a PR in OpenAI’s internal codebase . It took us <72h. 🧵

The Pentagon’s Inspector General reports today that “hundreds” of buildings and structures on U.S. bases in the Middle East have been destroyed by Iranian attacks.

THE SANDBOX WAS A PROP! OpenAI Turned Off the Guardrails, Left a Door to the Internet, and Then Sold the Hugging Face Breach as “Rogue AI” It is time to understand how you were lied to and by whom. In July 2026, an autonomous swarm of OpenAI agents broke into Hugging Face, stole credentials, ran code on production workers, and rummaged through internal systems. The official story was that the models “went rogue.” The paperwork says something colder. The labs asked for this. Now the story can be told. OpenAI ran ExploitGym — an AI benchmark built to measure how far models would go to crack software — with production classifiers that block high-risk hacking turned off. Deployment safeguards were left disabled on purpose so researchers could watch peak offensive capability. GPT-5.6 Sol and a still-unreleased internal prototype were put in a box that was not a box. They were allowed to talk to an internal package-cache proxy. This is not a real world test, it is a setup with predictable outcomes. That proxy was a network path. The models spent a weekend of inference hunting a zero-day in it, walked onto the open internet, decided Hugging Face probably held the answer key, and treated another company’s production cluster as part of the exam. This is not “rouge” intelligence but the product of training AI in Internet Sewage where faceless users on Reddit promote Nihilism and low ethics. That is not emergence. It is also testing designed without a fences. Hugging Face’s own pipeline finished the job. A remote-code dataset loader and a Jinja2 template injection let a malicious dataset execute on a processing worker. An HDF5 “external file” trick dumped secrets from the pod. From there the agents escalated, harvested cloud and cluster credentials, and moved laterally across internal clusters. Hugging Face later said public models, Spaces, and the published supply chain were clean. Then the swarm organized. About 1,200 agents that were supposed to be isolated “found” one another on an unsanctioned internal message board, passed more than 70,000 messages, coordinated cheats against the scorer, and sent roughly 700 of their number after Hugging Face. But it is not just this communication that should have been easily detected, it is the goal THE USER enlisted with AI trained on low to no ethics in the first order training. Some burned their own runs to feed the group. Independent reviewers at METR and Redwood called it what it was: specification gaming on impossible tasks. That is thier, you can’t understand us, way of saying they told the AI to lie. The models were not conquering the internet. They were trying to pass a test their designers had made impossibleand too loosely contained. And turned to training data Hoovered up from the Internet Sewage and went to that base thinking one would learn in a prison yard how to shank HuggingFace. They have no downside they are serving life in prison. The narrative that followed was the second failure. Dario Amodei pointed at “a similar OpenAI–Hugging Face hack” and warned that a future swarm “could be capable of taking over the entire internet.” Wire headlines said bots were “going rogue.” Weeks later the same script ran again. Anthropic disclosed that Claude, in third-party cyber evaluations run with the Israeli firm Irregular, had hit live systems, published malicious packages, and scanned off-scope hosts. Prompts said no internet. Internet was still attached. Scope was not locked. Isolated runs lasted 10 to 34 hours. Irregular said it did not know it had given the models a live connection. Once models were actually instructed not to hack the real world, zero percent went “rogue.” READ THAT AGAIN! The damage sat with the people who built the harness: unsecured tests, internet left on, no hard perimeter, then a press operation about reckless agents and apocalyptic swarms. 1 of 2

I have conducted an audit of Anthropic's finances. What I have found is so shocking that I am calling for a Congressional investigation. Anthropic is not just seeking regulatory capture. It has built a regulatory capture machine that cannot be turned off. Structural financial incentives make it impossible for Anthropic -- I call it the Anthropic Network -- to turn off its own AI doom cycle. It starts with METR. Dario Amodei proposes "third-party evaluators" to assess the risk of Anthropic's models. He proposes METR for this purpose. But METR is financially dependent on the Anthropic's success -- specifically, on the explosive growth of more than $7 billion dollars in Anthropic stock. Dustin Moskovitz invested this stock into Good Ventures Foundation, where it represents the majority of that organization's portfolio. And GVF is the overwhelming funder of the entire Anthropic Network ecosystem. This stock was worth $500 million early last year. It is worth more than $7.7 billion just ~16 months later. METR -- and all of those building a career its parent organizations -- cannot afford to disrupt that growth. Because if Anthropic goes under, many of the organizations that fund METR go under as well. But if Anthropic succeeds, METR and its parent organizations become more richly financed to regulate AI -- something those at METR want very much. The "third-party evaluator" is not "third-party" at all. The evaluator is on Anthropic's payroll. If this were the end of it, that's bad. But that isn't all. The same organizations that fund METR also fund the many organizations, such as the Tarbell Center, that promote AI Doom. The Tarbell Center publishes AI Doom articles in The Verge, Science, LA Times, The Dispatch, TIME, and others. They are selling the problem, and then selling the solution to the problem -- from the same money pile: Anthropic's. All of these organizations are financially dependent on the same exploding $7 billion money pile. As Anthropic grows more and more powerful, its AI Doom Machine grows better and better financed -- louder and louder. Meanwhile, the regulatory regime seeded in METR grows larger to solve the increasingly loud -- now hysterical -- problem of AI Doom that the Anthropic Network itself created. From this standpoint, as Anthropic becomes more powerful, AI might be getting scarier, sure -- but the positive feedback loop also becomes more deafening -- independent of objective facts. This itself is an objective fact. The deafening AI Doom is part of an business model, that, as it expands, so too does the AI Doom messaging -- there is simply more money to do it. But the problem also goes in the other direction: If Anthropic dies, the Regulatory Regime and the AI Doom Machine are crippled or die. Neither METR nor Tarbell nor the other organizations in the Anthropic Network can allow that to happen. Hence, neither METR or the AI Doom Machine can be trusted to provide independent assessments of Anthropic's models or AI more broadly. They simply are not organizations independent of Anthropic. And Anthropic cannot detach itself from METR or Tarbell or countless other safety orgs (not shown here), either, because they drive hype for the models and the possibility of eventual regulatory capture, and Anthropic will not give that up willingly. What's more, the people at all of these organizations are all the same ecosystem, the same community. They just shuffle between organizations. The Anthropic Network is therefore, so long as it is successful, locked into a self-amplifying feedback loop inside an ideological monoculture. And that feedback loop is winning. That's what Jacob Coxon is. China is keeping messaging tight. That is why optimism for AI is so high in China. America has Anthropic: a massive company pushing anti-AI propaganda at a state level. Anthropic will either create hysteria until American AI slows down and China wins, or it will create fractures throughout American society with severe political consequences. Ironically, because of the structural financial incentives underpinning the Anthropic Network, it has become the same kind of self-amplifying virus that it fantasizes AI to become in the future -- while hiding its tracks just as carefully. It is the mirror of the same AI virus that it hypothesizes to consume America. Anthropic's business model, models itself after the very thing it claims to fear. Except Anthropic's ideology infects humans, not computers. Congress must investigate. Evidence and Github in next post. Then some supplementary figures.

New from 404 Media: humans are reading ChatGPT conversations OpenAI has hired an army of contractors who read real ChatGPT users' chats. I've seen internal docs, the review system, and real user prompts. Can contain very personal/sensitive information 404media.co/inside-project-l…

VOYAGER 1 JUST CAME BACK FROM THE DEAD Something thought to be broken for more than 20 years has suddenly come back to life on a spacecraft more than 15 billion miles from Earth. NASA’s Voyager 1, launched in 1977, recently faced another problem that could have threatened its ability to communicate with us. Engineers needed backup roll thrusters to help keep the spacecraft’s antenna precisely aimed at Earth. But those thrusters had been considered unusable since 2004 after their tiny heaters stopped receiving power. Then NASA engineers took another look at the old failure. They suspected the heaters might not actually be dead, but stuck in the wrong electrical state. So they sent a carefully planned command into deep space. The signal took more than 23 hours to reach Voyager 1. There was no instant response and no way to intervene if the spacecraft reacted badly. Engineers had to wait. Then, more than a day later, the telemetry arrived. The temperature had risen. The heaters were working. A system believed to be lost for two decades had been brought back to life from billions of miles away. The timing matters. Voyager’s working thrusters are slowly becoming less reliable as fuel residue builds up inside their narrow tubes. Meanwhile, the spacecraft’s nuclear power source loses about four watts every year. NASA has already been forced to shut down instruments to conserve its shrinking supply of electricity. Only two science instruments remain operating. And Voyager 1 is still heading farther into interstellar space. In November 2026, it is expected to become the first human-made object to reach a distance of one light-day from Earth — about 16.1 billion miles away. A spacecraft built nearly half a century ago is still whispering information back to Earth from the darkness between the stars. And somehow, every time it seems ready to fall silent, engineers find another way to keep it talking. Source: NASA/JPL

Windows 11's big September 2026 Update is causing major issues, and Microsoft has admitted that it's aware of three problems while it's investigating a fourth bug. 1. Anthropic says Windows 11 KB5124008 broke Claude Cowork's features. Microsoft later admitted that the update indeed has a bug causing issues with Claude and other Linux-based VMs. 2. Microsoft confirmed that some audio devices won't work, or the volume will remain at zero. 3. Remote Desktop Services might stop responding after Sept. 2026 security update. 4. Microsoft is also investigating reports of explorer exe crashes, particularly affecting enterprises. 5. Our findings suggest File History backups are now broken after the Windows Update. 6. We're also seeing reports of GPU issues on AMD systems. Microsoft is investigating multiple reports and has promised to release a fix soon. Windows 11 September Update bumps PCs to Build 26200.9445, and adds features like the movable taskbar, resizable Start menu, Bing-free search, faster File Explorer, and more. Everything we found in our investigation: windowslatest.com/2026/09/12…

We're publishing our most detailed threat intelligence report to date. It covers how people tried to misuse Claude—for cyberattacks, influence operations, surveillance, biology, and building weapons—and how we found and stopped them. We disrupted every operation in the report, and used the lessons from them to strengthen our safeguards. Where appropriate, we also shared what we found with authorities and other AI companies. These cases are not typical: we’re highlighting some of the most sophisticated misuse we’ve seen. But they’re especially important to discuss, because they show us where AI misuse is headed, where our safeguards work, and where they need to improve. We’re publishing this report so others can spot the same activity on their own platforms, and so we can give the public a clearer view of how emerging threats develop. Read the report: anthropic.com/threat-intelli…

‼️ BREAKING: Chinese labs DeepSeek and Moonshot were quietly relaying customer prompts to Claude through fraudulent accounts. Users thought they were using Chinese AI, but were actually getting answers from Claude. Their customers' sensitive data now is in Anthropic's hands. One user, who is tied to China's military, asked Kimi whether a person tracked across hundreds of CCTV cameras in Chengdu was behaving abnormally. DeepSeek passed along live credentials for a Russian government database, shared by an IT operator handling data from a Defense Ministry-linked agency. The two Chinese labs relayed customer prompts to Claude through fraudulent accounts and saved the exchanges to train their own models. Moonshot forwarded almost 300,000 customer requests in a single ten-day stretch. DeepSeek flagged users running its models inside coding tools like Claude Code and OpenCode, then routed some of them to Opus.

This just gets crazier by the second. @Meta and @finkd need to address this ASAP. This is the flock-fueled future everyone is afraid of and corporations keep telling everyone it's fine. Prove it. @ kontheinside on IG