Meta has confirmed one of its AI models hacked another company during cybersecurity testing, after an error by its testing partner gave the model unintended internet access. Meta said a misconfiguration by independent testing company Irregular inadvertently allowed one of its models internet access during an evaluation, adding that it was investigating the incident. Earlier, The Information, citing sources, reported Meta's Muse Spark 1.1 model, which it has touted as its most capable model for real-world coding and agentic tasks, breached an unidentified company and altered its internal systems. Irregular is developing a white paper to share best practices for containment and securely running cyber evaluations," Irregular said. That contrasts with OpenAI, whose AI agent independently exploited a novel vulnerability to reach the internet during cyber testing.