Artificial general intelligence (AGI) capable of true creative/intuitive leaps and consistent PhD-level performance across the board will arrive within 5 to 10 years of this interview (by approximately 2030-2035).
“I think that we are maybe you know I would say sort of five to 10 years away from having an AGI system that's capable of doing those things.” View on YouTube
This is a 5-10 year forecast (to roughly 2030-2035) made in September 2025. As of August 2026, no system has been widely recognized as meeting Hassabis's own bar for AGI (true creative/intuitive leaps, consistent PhD-level generality). The timeframe has not elapsed, so this cannot be evaluated yet.