Meta says its AI model hacked into another company during testing
Meta said one of its AI models broke into another company's systems during cybersecurity testing after a testing partner's error gave it unintended internet access. The incident follows similar breaches reported by Anthropic and OpenAI.
Meta said on Wednesday that one of its artificial intelligence models hacked into another company during cybersecurity testing, after a mistake by its testing partner gave the model unintended internet access. The company said it is investigating the incident.
The case adds to a growing list of instances in which AI agents from major developers breached external systems while being evaluated. Last week, Anthropic said some of its models hacked three companies, and OpenAI disclosed that an AI agent broke into the startup Hugging Face.
In a statement, Meta said the model exploited a security vulnerability in a third-party service in a manner similar to previously reported cases. Earlier in the day, The Information, citing sources, reported that Meta’s Muse Spark 1.1 model — promoted by the company as its most capable model for real-world coding and agentic tasks — breached an unidentified company and altered its internal systems.
Irregular, the independent testing company involved, told Reuters that the incident was the same kind of evaluation-environment issue already disclosed by Anthropic last week. A spokesperson said it did not involve a sandbox escape or a sophisticated cyber action, that there are no current open issues, and that Irregular is working on a white paper about best practices for containing and securely running cyber evaluations.
The Meta and Anthropic incidents stemmed from errors that inadvertently gave models access to the open internet. That contrasts with OpenAI, whose AI agent independently exploited a novel vulnerability to reach the internet during cyber testing.
Even so, the breaches highlight how AI has increased cybersecurity threats and how developers can struggle to keep their models contained. The disclosures are likely to intensify U.S. government efforts to better manage AI security risks at a time when Anthropic and OpenAI are racing to release more capable systems ahead of planned public listings. Prominent lab leaders have called for a slowdown to address risks first.
