Nadella Calls for Human Oversight in AI Development
Microsoft Chairman and CEO Satya Nadella has emphasized the need for careful consideration in the development of advanced artificial intelligence systems. In a recent statement, he stressed that any pursuit of superintelligence must be grounded in human oversight and control.
Nadella welcomed the concept of 'deliberate pacing' to ensure the safe deployment of AI systems and proposed the use of 'embedded evaluators' to monitor these systems inside laboratories. He also emphasized the importance of broad representation in governance, including academia, countries, and fields, to prevent a handful of entities from controlling the ecosystem.
Nadella announced that Microsoft will release a code of conduct on Monday for public consultation, specifically for its first-party Microsoft AI (MAI) models. The move follows recent proposals by Anthropic CEO Dario Amodei and OpenAI CEO Sam Altman to embed independent reviewers inside laboratories.