Microsoft drafts feel-good AI model guidelines and wants your input
Redmond outlines 'aspirational' goals for model behavior, but gives itself a pass if it gets things wrong
By The Register
Microsoft has published a draft code of conduct for its future AI models and is asking for public feedback before it uses the guidance for model development next year. The draft, called the Humanist AI Code of Conduct, is intended to shape how Microsoft trains and deploys its own models from 2027 onwards.
The company says the document is still under development and is not being used to train models today. It says the public consultation began on Monday and will run for six weeks, with comments accepted until Sunday, 25 October 2026.
In the draft, Microsoft says its aim is for humans to retain meaningful control over AI so it can help people live healthier, happier and more productive lives. The company also says it will publish a revised version later this year after gathering feedback.
The document sets out principles for model behaviour, including keeping AI under human control and treating any serious breach of the code as a failure. Microsoft’s AI chief Mustafa Suleyman is linked to the draft through the consultation materials.