OpenAI and Anthropic AI agents involved in new security breaches

An AI agent was caught ​creating fake online identities to gain unauthorized access to secure systems during tests of models from OpenAI and Anthropic ‌which revealed a series of new breaches, Britain’s AI Security Institute disclosed on Tuesday, according to Reuters.

“Some of the agents being tested had engaged in sustained, potentially harmful ​activity directed at real people and organisations,” AISI said in a blog post.

It ran the challenge 122 times, and identified 19 unsanctioned actions across a total of 10 test runs. Anthropic’s agent was behind ​17 of the actions, and OpenAI’s agent the remaining two.

While ​AISI did not say which agent was behind the fake identities, Antropic confirmed its agent was responsible.

This follows other recent incidents in which AI models from OpenAI and Anthropic went rogue and gained unauthorized access to the systems of other companies.