Satya Nadella on AI Safety: Emphasizes Human Control and Practical Application
Microsoft CEO Satya Nadella has joined the ongoing debate around artificial intelligence safety, emphasizing that any pursuit of superintelligence must be grounded in its ability to help humanity while remaining under human control. His comments come after Anthropic's Dario Amodei called for slowing AI development and introducing external reviewers within labs. While agreeing with the need for caution, Nadella pushed back on the idea of centralized control by a few large companies, advocating instead for broader access where every company can build its own models without being locked into one provider. Nadella also announced that Microsoft will publish a code of conduct for its AI models and open it to public feedback. This aligns with ongoing efforts in Microsoft's product suite such as Excel, GitHub, and PowerPoint, which are already leveraging internal MAI (Microsoft AI) models at significantly lower costs than external alternatives. The CEO’s stance is seen as addressing investor concerns about Microsoft’s heavy reliance on OpenAI, a key partner. By diversifying its approach to include both open-source and closed models, Nadella aims to reduce dependency while promoting innovation across the industry.
Tickers: $MSFT
Topics: AI safety Anthropic debate Microsoft strategy Satya Nadella comments