#AI code of conduct

Microsoft Wrote Down What Its AI Can and Cannot Do in a Cyberattack

A newly published set of rules governs how Microsoft's AI models may be used in security work, separating legitimate defence research from operational attack capability.