Microsoft CEO Nadella Outlines AI Development Vision, Backs Independent Audits and Announces Conduct Guidelines

Microsoft CEO Satya Nadella on Monday shared the company's vision for frontier AI in a post on X, stressing that any pursuit of superintelligence must operate under core principles. He said that if people build AI that cannot help humanity and cannot be controlled by humans, then it is not worth developing.

Nadella argued that AI needs to accelerate so its benefits can reach different countries, communities and businesses. He said a thriving frontier ecosystem requires both open-source and closed-source models. For enterprises, he noted, companies should fully own their tacit knowledge and be able to build systems with continuous learning loops, iterating without depending on a single model vendor.

Microsoft said it welcomes genuine AI alignment research and the idea of independent audits, with plans to put these concepts into practice. Nadella added that these principles should not be controlled by a few companies, but should involve people from different ecosystems, countries and domains.

The CEO also outlined Microsoft's approach, including broad access and choice at every layer of the AI stack, letting enterprises control their own learning loops and models, and a code of conduct supporting Microsoft's in-house MAI models. He said the guidelines will be released on September 15 for public feedback.