Microsoft releases humanist AI code of conduct emphasizing people over AI

Microsoft announced today the release of a 37‑page “humanist AI code of conduct,” a set of principles that emphasize that “people matter more than AI,” deny that AI systems are conscious, and reject the notion of granting legal personhood to artificial agents. The code, which also states that AI models should not be designed to imitate consciousness, comes as industry leaders grapple with mounting safety concerns about the rapid advancement of increasingly complex AI models.

The timing of the announcement follows recent warnings from AI researchers that development is outpacing the ability to safely deploy, verify, and control such systems. Anthropic CEO Dario Amodei called for a coordinated slowdown of AI progress over the weekend, urging the sector to address the risks before further breakthroughs. Microsoft’s new conduct guidelines aim to provide a framework for responsible development while the broader community debates how to balance innovation with safety.