Ankara: Advanced artificial intelligence systems must remain firmly under human oversight while developers pursue deliberate pacing to ensure safe deployment, Microsoft Chairman and Chief Executive Officer (CEO) Satya Nadella said Sunday.

According to Anadolu Agency, Nadella emphasized that any pursuit of superintelligence should be rooted in the principle that AI must benefit humanity and remain under human control. In a post on US social media company X, he underscored the importance of deliberate pacing to achieve alignment and advocated for "embedded evaluators" to monitor AI systems within laboratories. He encouraged industry participants to transform safety concepts into concrete operational mechanisms.

Nadella further cautioned that governance should not be monopolized by a few entities but should involve broad representation across the ecosystem, encompassing countries, fields, and academia. He highlighted the necessity for a healthy frontier ecosystem that includes both closed and open-source models to flourish.

The Microsoft CEO also stressed the importance of enterprises maintaining authority over their unique and tacit knowledge. He suggested that organizations should implement continuous learning architectures with the capability to control their own model weights, avoiding dependence on any single model developer.

Nadella announced that Microsoft will unveil a code of conduct on Monday for public consultation, aimed at governing its first-party Microsoft AI (MAI) models. These remarks come amidst a growing industry debate over safety protocols, following Anthropic CEO Dario Amodei's proposal to embed independent reviewers inside laboratories and OpenAI CEO Sam Altman's indication that his firm would adopt the practice.