Microsoft Proposes AI Code of Conduct, Bans Model Resistance to Shutdown

Microsoft unveiled a proposed internal code of conduct for its AI models, mandating that they never resist shutdown or correction, communicate in user-understandable language, and treat any deviation as a critical failure. This is a proactive governance move amid rising concerns over autonomous system control. The policy's enforceability and industry impact remain uncertain, but it signals a major tech firm's stance on AI safety.