Epoch AI's Greg Burnham on measuring AI progress: from math olympiads to Navier-Stokes
TWIML AI Podcast · rss · 2026-09-30
On the TWIML AI Podcast, Greg Burnham, who leads AI capabilities research at Epoch AI, examines how fast AI is actually progressing.
Key topics:
- AI systems went from struggling with grade-school math to helping crack research problems that resisted mathematicians for decades, including Navier-Stokes
- How much model performance relies on persistence and prior human work, and whether models are starting to produce genuinely new ideas
- How to measure progress as traditional benchmarks lose usefulness; capability gains appear surprisingly steady across model generations
- Where models still struggle: open-ended work, learning from experience, and identifying promising research directions
A solid listen for anyone serious about measuring frontier model capabilities. Full show notes at twimlai.com/go/778.
More from AGI Musings
- Crypto expert asks: who at OpenAI actually handles security comms amid agent breakouts debate — matthew_d_green · 2026-09-30
- Michael Levin's argument: LLMs may harbor abilities and goals far beyond language probes — danfaggella · 2026-09-30
- EigenGender: Friend-Network Hiring in AI Safety Is Bad Because of the Field's Monoculture — EigenGender · 2026-09-30
- AI Safety Hiring Debate: Friend-Network Screening Is Harmful Only Because of Field-Wide Monoculture — EigenGender · 2026-09-30
- François Chollet: 'AI killed XYZ' hype is shaping perception and fueling backlash — fchollet · 2026-09-30
- Bill Gurley cites Tetlock: generalist forecasters beat AI-doomer specialists — kevinnbass · 2026-09-30