Microsoft published its first formal Code of Conduct for its in-house MAI models on September 14, 2026, establishing outright bans on uses such as cyberattacks, nuclear weapons development, and deepfake production. CEO Satya Nadella framed the move as part of an argument for decentralized AI governance, saying “AI governance must not be controlled by a handful of entities.” The company opened the draft, developed over roughly five to six months, for six weeks of public feedback before finalizing it.
