Microsoft CSO Eric Horvitz Rejects AI Pause, Proposes Safeguards
Eric Horvitz, Microsoft’s Chief Scientific Officer and a leading artificial intelligence expert, argues against the proposed six-month pause on advanced AI development advocated by figures like Elon Musk. In a recent interview, Horvitz stated that a temporary halt would not resolve the underlying risks associated with rapidly evolving large language models. Instead, he emphasizes the importance of responsible AI deployment, rigorous internal testing, and limited public previews to identify potential failures. Horvitz highlights that real-world interactions often reveal vulnerabilities that internal red-teaming might miss, citing incidents where users intentionally pushed systems to their limits. As a member of President Biden’s Council of Advisors on Science and Technology, Horvitz is actively involved in shaping policy and ethical guidelines for AI. He acknowledges the danger of bad actors weaponizing AI but believes in empowering humans through technology while maintaining strict oversight. His approach focuses on understanding the complex nature of intelligence and ensuring that AI systems remain aligned with human values and safety standards, rather than stopping progress entirely.
Wire timeline
Microsoft CSO Eric Horvitz Rejects AI Pause, Proposes Safeguards
Eric Horvitz, Microsoft’s Chief Scientific Officer and a leading artificial intelligence expert, argues against the proposed six-month pause on advanced AI development advocated by figures like Elon Musk. In a recent interview, Horvitz stated that a temporary halt would not resolve the underlying risks associated with rapidly evolving large language models. Instead, he emphasizes the importance of responsible AI deployment, rigorous internal testing, and limited public previews to identify potential failures. Horvitz highlights that real-world interactions often reveal vulnerabilities that internal red-teaming might miss, citing incidents where users intentionally pushed systems to their limits. As a member of President Biden’s Council of Advisors on Science and Technology, Horvitz is actively involved in shaping policy and ethical guidelines for AI. He acknowledges the danger of bad actors weaponizing AI but believes in empowering humans through technology while maintaining strict oversight. His approach focuses on understanding the complex nature of intelligence and ensuring that AI systems remain aligned with human values and safety standards, rather than stopping progress entirely.
Fortune | FORTUNE