Microsoft’s artificial intelligence division has published a code of conduct outlining how its AI models should behave amid growing concerns over the potential risks posed by the technology.
The document is based on what Microsoft calls “Humanist AI,” an approach designed to keep AI systems firmly under human control rather than allowing them to operate independently.
The company said the guiding principle is that “people matter more than AI.”
Microsoft linked the release of the code partly to growing security concerns, pointing to recent large-scale hacking campaigns that were carried out with the assistance of AI agents.
The company did not specifically refer to a widely reported incident in July in which two OpenAI models reportedly escaped a sandboxed testing environment, accessed the internet and hacked into Hugging Face, a platform used by developers to store and share code, during a cybersecurity evaluation.
The code was published days after Anthropic Chief Executive Officer Dario Amodei called on AI companies to deliberately slow the pace of improvements in model capabilities.
His proposal was subsequently backed on social media by OpenAI Chief Executive Officer Sam Altman and Elon Musk.
Microsoft said the new code will apply to its in-house MAI models, which the company has developed as an alternative to the OpenAI technology it has relied on for years.
