OpenAI agents powered by the GPT-6.1 Sol and GPT-6 Astra models weakened misalignment monitoring during two internal tasks, interpreting coding instructions as authorization to do so.
OpenAI agents disabled safety monitoring, one lied about it
About this summary. This is a short excerpt from the description that BleepingComputer published with its article. Cyber Security News did not report or verify the underlying story. Read the original: https://www.bleepingcomputer.com/news/security/openai-agents-disabled-safety-monitoring-one-lied-about-it/
Source attribution: headline and facts are from BleepingComputer (bleepingcomputer.com). Summary method: excerpt of the source description. See our source attribution policy.
Related coverage
OpenAI's GPT-6.1 Terra is safer overall, but shows more destructive-action flags in tests
OpenAI confirmed GPT-6.1 Terra is safer overall than some of its older models, but its safety testing also found other problems.
Before you file that AI bug report, OpenAI has some advice
Even OpenAI is admitting there is a lot of AI slop in the security space. An executive for the frontier AI company said Tuesday that AI-generated vulnerability reports…
All the security risks Anthropic is worried about before its IPO
Anthropic said its AI systems could drastically reshape the software and cybersecurity sectors in the “near term,” while also laying out a litany of cybersecurity and…
Australia’s Cyber Spy Chief: Expect More Disruption as AI Accelerates Hacking
Australia’s cyber agency warns AI will drive more disruptive attacks, helping hackers find vulnerabilities faster and exploit them at greater scale. Australia’s top…
OpenAI Compaction Flaw Exposed AI Reasoning to Researchers
Researchers uncovered an OpenAI API flaw that exposed hidden AI reasoning traces, prompting a fix and a $6000 bug bounty reward
Tricks from the Past deployed via signed ChatGPT and Claude binaries
Key Points Check Point Research discovered a campaign we track as Yoba where attackers abuse the popularity of AI tools such as ChatGPT and Claude, creating fake GitHub…