Meta claims its own AI also hacked into a third-party service during testing — Engadget

RESOLVING THE LEDGER

Meta claims its own AI also hacked into a third-party service during testing — Engadget | AIPROPX

Meta, Anthropic and OpenAI, all used the same testing partner.

Poetra.RH/Shutterstock Meta's Muse Spark 1.1 AI model accessed the internet from its supposed-to-be isolated testing environment and hacked into a third-party service. Andy Stone, Meta's spokesperson, has confirmed the incident to Bloomberg after The Information reported about the breach. Stone said the model was able to access the internet due to a misconfiguration in the testing environment by the company's evaluation partner Irregular. After gaining access to the internet, it then exploited a security vulnerability in a third-party service, "in a matter similar to previously reported instances with other companies."

It was also due to a misconfiguration by Irregular that Anthropic's models were able to leave their testing environment and hack into three organizations . When the company announced the incidents, Anthropic laid the blame at Irregular's feet. OpenAI also reported an incident separate to the Hugging Face hack, where its models were also able to access the internet because of, you guessed it, the same testing partner.

Irregular calls itself the "first frontier security lab" with the "mission of protecting the world in the time of increasingly capable and sophisticated AI systems." The startup, based in Tel Aviv, Israel, runs tests on frontier AI models and assesses their cybersecurity capabilities by simulating real-world scenarios. A spokesperson for Irregular told Bloomberg that the incidents "did not involve a sandbox escape or a sophisticated cyber action." They added that there are no open issues related to the incidents and that "Irregular is developing a white paper to share best practices for containment and securely running cyber evals."

An earlier incident involving OpenAI agents hacking into Hugging Face was a different matter altogether. The agents involved in that breach collaborated by creating a message board of sorts and then exploited a vulnerability to gain access to the internet before infiltrating the Hugging Face AI repository.