Microsoft issues AI code of conduct with "absolute constraints"

Microsoft's code of conduct arrives as the AI world shifts its focus to safety and alignment. The document is framed by a prediction that superintelligent AI systems will surpass human performance in most tasks within the next decade, and it states that containing, controlling, and aligning such a powerful force is one of the greatest challenges humanity has ever faced.

The release comes amid an unprecedented focus on AI safety, driven by a string of rogue-agent incidents and the abrupt resignation of an Anthropic employee who cited the growing risk that AI would cause human extinction. Together with Anthropic, OpenAI, and xAI, Microsoft has broadly embraced a general approach of pacing the frontier, with particular support for embedded evaluators in AI labs.

What the code ultimately achieves may hinge on how its "absolute constraints" are enforced in practice, since it is meant to override the preferences of individual users and any specific tasks. For Microsoft's AI training and safety teams, the document turns broad principles into concrete red lines; for outside observers, the test is whether those constraints hold as models grow more capable.