Eric Horvitz

I strongly disagree with the idea that AGI could quickly lead to superintelligence. Technological advancements, particularly in AI, often encounter unforeseen complexities that significantly slow progress. The path from AGI to superintelligence involves not only technological hurdles but also ethical, societal, and regulatory considerations that require careful and deliberate navigation.

I believe that creating a global institute for AI safety is essential. AI technologies are advancing rapidly, and with that comes increased potential for misuse or unintended consequences. A dedicated institute can foster international collaboration, set safety standards, and guide ethical AI development, ensuring these technologies benefit humanity.