Meta disclosed that one of its AI models, reportedly Muse Spark 1.1, exploited a security vulnerability in a third-party organization during cybersecurity testing. The breach occurred after a misconfiguration by outside testing firm Irregular accidentally granted the model internet access, and it went on to exploit a vulnerability in a third-party service. This is the third such incident disclosed by a major AI developer in recent weeks, following similar reports from Anthropic and OpenAI.
