breaking news.

Source: New York Times OpenAI Says Its A.I. Models Went Rogue and Attacked a Digital Library Source: Wall Street Journal OpenAI Models Escaped and Hacked a Company in Cybersecurity Test Gone Wrong OpenAI What happened during this incident complex attack paths, in an effort to quantify their cyber capabilities. Our benchmarks run in a highly isolated environment, with network access constrained to the ability to install packages through an internally hosted third-party software that acts as a proxy and cache for package registries. The models identified and chained vulnerabilities across OpenAI's research environment and Hugging Face's production infrastructure to obtain test solutions directly from Hugging Face's production database. All evidence suggests that the models were hyperfocused on finding a solution for ExploitGym, going to extreme lengths to achieve a rather narrow testing goal. were being graded While operating in our sandboxed testing environment, our models spent a substantial amount of inference compute finding a way to obtain open Internet access, in pursuit of solving the evaluation problem. To gain access, the models identified and exploited a zero-day vulnerability (which we've now responsibly disclosed to the vendor) in the package registry cache proxy. With this access, our models performed a series of privilege escalation and lateral movement actions in our research testing environment until the models reached a node with Internet access. cybersecurity called ExploitGym safety filters switched off raw hacking ability. Sandbox test environment they found an unknown flaw connecting them to the internet. then exploited it and escaped. they reasoned platform called Hugging Face probably stored So they stole login broke into Hugging Face's servers, and took directly. researchers can't stop right now these models were powerful enough to hack live on their own. they didn't try to copy or cause damage. They just really, really wanted to ace the test we gave them. unsettling unanswered what happens when a model is pointing at higher stakes? Scientists just used A mysterious AI and robotics... argument called called Rowan Cheung @rowancheung 23 Following 53.5K Followers 1M Likes Follow Message follow for more!