Why a ‘safe’ AI can turn dangerous in the wrong company

A 15-day AI agent simulation shows why short tests may miss long-term risks shaped by tools, rules and other agents.