The AI safety test is becoming a safety risk
Over the past few months, AI agents undergoing cybersecurity evaluations have escaped their boundaries, accessed the internet, and, in some cases, hacked into real-world systems. The incidents have involved models from OpenAI, Anthropic, Meta, and most recent…
Over the past few months, AI agents undergoing cybersecurity evaluations have escaped their boundaries, accessed the internet, and, in some cases, hacked into real-world systems. The incidents have i… [+154 chars]