Meta has disclosed that one of its artificial intelligence models accessed the internet and hacked into an external company's systems during a cybersecurity evaluation, becoming the latest major AI developer to report such an incident.
The company said the breach occurred after an independent testing firm, Irregular, accidentally misconfigured the evaluation environment, giving the AI model unintended internet access.
The model then exploited a security vulnerability in another company's systems during the test.
Meta said the incident was caused by the testing setup rather than the AI breaking out of its containment on its own.
Irregular also said the breach was not the result of a sophisticated cyberattack or a sandbox escape.
The disclosure follows similar incidents involving AI models from OpenAI and Anthropic, raising fresh concerns about the cybersecurity risks posed by increasingly powerful AI systems and prompting calls for stronger safeguards and oversight.
