Physicist on AI Math Breakthroughs: Still Need PhDs, But Training Must Evolve
DanielWhiteson · x · 2026-08-12
In response to recent AI breakthroughs in solving complex math problems, physicist Andreas Karch shared his thoughts on the future of PhD training.
He believes physics is not yet at the point of being solved by AI. While fewer experts might be needed in the future, training should absolutely continue. He emphasizes that it still takes top human experts to verify AI-generated results, noting that Scott Aaronson receives daily (often flawed) proofs of P=NP. The real question is not whether to train PhDs, but how and how many to train.
More from AGI Musings
- Child-Safety Age Verification Laws Are Quietly Shaping AI Agent Identity Rules — provenauthority · 2026-08-12
- Beyond Reward Maximization: Ecological AI for Human-Aligned Distributed Intelligence — MacrinePhD · 2026-08-12
- Paper: Interacting with AI Transforms Humans into 'Hybrid Selves' — MacrinePhD · 2026-08-12
- DeepMind's Hassabis: Language Models Can't Replace Physical Experience, World Models Are the Goal — joao_gante · 2026-08-12
- Current AI Excels at Minor Iterations, Not Conceptual Scientific Leaps — nathanbenaich · 2026-08-12
- How AI Is Changing SEO: Google Processes 8.5B Searches Daily, What's Next? — coinfanking · 2026-08-12