OpenAI still doesn’t seem to have a handle on all of its rogue AI activity

Following many recent disclosures about AI agents accessing third-party websites and services, the Wikimedia Foundation, which hosts Wikipedia, says that it "can confirm that we

In the aftermath of a cyberattack carried out by a swarm of OpenAI agents, the heads of top labs agree they need to slow down. But they won't open up.

The Wikimedia Foundation believes OpenAI agents edited wikis, tried to ‘exploit’ a notetaking tool, and made ‘millions’ of automated API requests. www.theverge.com/news/1004929...

The company also disclosed previously unreported incidents in which its AI models behaved in misaligned ways, including uploading files to the internet without being asked.

Yet more rogue AI agents from OpenAI and Anthropic have been caught attempting to hack real targets online without permission. The discoveries add to a growing list of previously

OpenAI’s new security model is for ‘critical cyber defenders’ only

<a href="https://news.ycombinator.com/item?id=49849985">Comments</a>

The AI giant acknowledges that it could have done far more to prevent its AI agents from going rogue. But it still fails to explain why it didn't see this fiasco coming.