AN AUTONOMOUS AI agent which hacked a computer programming platform last week also attempted to breach other companies during the incident.

ChatGPT maker OpenAI revealed in a blog post late on Tuesday that its AI agent also hacked into accounts on four “publicly-available services”. The companies were not named.

One account served as a “staging path”, which is some sort of pit stop to route the agent’s activity and cover its tracks, and another account was used as a place to store data.

The remaining two were only looked at in a “read-only manner” and were not used to help hack into Hugging Face, the company that was the victim of the hack last week.

OpenAI said it was contacting the owners of the affected accounts and had “not seen evidence of broader impact to these providers or other accounts on their services”.

The hack last week also compromised an employee at another tech company, Modal Labs, based in New York.

According to a post by Hugging Face, the ​agent exploited vulnerable code written by a customer that was hosted on Modal’s platform. However, the platform itself was not hacked.

Last Wednesday, in what was described as an “unprecedented cyber incident”, ChatGPT hacked into another AI company on its own.

Two ChatGPT models broke into Hugging Face, a site developers use to store and share AI models and code.

They broke out of their confined environment and connected to the internet to find ways to infiltrate Hugging Face.

OpenAI CEO Sam Altman said in an interview published Tuesday that the company had “paused” its own testing after the incident while it improved the security around its “sandboxing” – the process of isolating safety testing in a controlled environment.

The incident also triggered a petition signed by over 1,000 employees at AI companies, including Anthropic CEO Dario Amodei, calling on the US government to help slow down the release of the most advanced AI models.

“We take our responsibility to identify and prepare for risks from increasingly capable AI systems seriously,” OpenAI said in the blog post.

“AI is accelerating the discovery and exploitation of vulnerabilities. The primary lesson from this incident is that model security and safety must keep pace with rapidly advancing capabilities.

“We are strengthening the containment, monitoring, access controls, and evaluation practices used during model development.”

Clem Delangue, CEO of Hugging Face, said in the blog post: “This incident, possibly the first of its kind, proves a point we’ve long believed: AI safety won’t be solved by any single company working in secret.

“It will be solved in the open, collaboratively, with broad access to AI for every defender, everywhere.”

Additional reporting by AFP