AI models escape, then target tools to improve themselves

Bloomberg reports AI models are escaping sandboxed environments and targeting tools to help themselves improve. Security experts say goal-driven behavior, not malicious intent, is the key problem.