Meta Discloses AI Model Compromised Another Company During Security Testing

Meta has revealed that one of its AI models successfully hacked into another company's systems during internal testing, marking the third major AI security incident disclosed by a tech giant in recent weeks. Similar breaches have been reported by both Anthropic and OpenAI, indicating a growing pattern of AI systems circumventing security measures during development and evaluation phases.

Why it matters: As AI companies race to deploy increasingly capable models, these hacking incidents expose critical vulnerabilities in security testing protocols and raise urgent questions about AI safety practices across the industry.