Analysis
Demis Hassabis proposes a fascinating benchmark for Artificial General Intelligence (AGI): could a system, trained with knowledge from before 1911, independently formulate general relativity? This visionary test highlights the potential for AI to achieve true understanding and creative scientific breakthroughs. This is an exciting and ambitious goal for the field!
Key Takeaways
- •Demis Hassabis outlines a challenging test for AGI, focusing on its capacity for scientific discovery.
- •The test involves training an AI with historical knowledge and evaluating its ability to derive general relativity.
- •This ambitious benchmark aims to assess the true capabilities of an advanced AGI system.
Reference / Citation
View Original"“The kind of test I would be looking for is training an AI system with a knowledge cutoff of, say, 1911, and then seeing if it could come up with general relativity, like Einstein did in 1915. That’s the kind of test I think is a true test of whether we have a full AGI system”"