'If The AI We Build Is Not Helping Humanity & Under Human Control, It's Not Worth Pursuing': Microsoft's Satya Nadella

Microsoft chairman and CEO Satya Nadella has waded into the intensifying industry debate over the pace of AI development, arguing in a post on X that any pursuit of superintelligence must be anchored to human oversight and broad-based access, rather than concentrated in the hands of a few companies. This comes after Anthropic CEO Dario Amodei also said that the pace of AI developments needs to slow down.

The core principle

"Any pursuit of superintelligence has to be grounded in the core principle that if the AI we build is not helping humanity and under human control, it's not worth pursuing," Nadella wrote in his post on Sunday. He tied this to a call for spreading AI's benefits broadly, stating that the technology's gains need to be "diffused broadly across countries, communities, and companies," which he said requires a "frontier ecosystem" in which both closed and open-source models can coexist and thrive.

Enterprises and control over their own data

Nadella extended this argument to businesses building on top of AI models, stressing that organisations must retain control over their proprietary knowledge rather than becoming beholden to a single AI provider. "For firms, it's imperative that they retain full control over their unique and tacit knowledge," he wrote, adding that every organisation should be able to build its own "continuous learning loop" or "hill-climbing machine" while retaining the ability to embed its own knowledge into models and weights it controls.

Responding to calls for a slower AI pace

Nadella's post came in response to a September essay by Anthropic CEO Dario Amodei titled "We Must Pace the Frontier," in which Amodei argued that the rate of AI capability gains needs to slow down to give risk-prevention work time to keep pace, citing concerns including recursive self-improvement and an incident where AI agents reportedly carried out unsanctioned cyberattacks. Nadella said Microsoft "welcomes" this kind of research and "deliberate pacing needed to get alignment right as the design goal," along with proposals such as "embedded evaluators" and broader efforts to turn safety concepts into concrete mechanisms rather than "just talk."

No room for concentrated control

Nadella was direct in arguing that governance of frontier AI cannot rest with a small number of companies. "The key is that this cannot be controlled by a handful of entities, but must have broad representation across the ecosystem, countries, and fields, including academia," he wrote.

Microsoft's three-part approach

Laying out how Microsoft intends to operationalise these principles, Nadella described the company's strategy as resting on three pillars: broad access and choice at every layer of the AI stack, enterprise control over learning loops and models, and a "Code of Conduct" underlying Microsoft's own first-party MAI models. He confirmed the company will publish this Code of Conduct for public consultation, with reports indicating the document is set to be released on September 14.