Claude's Breakthroughs in String Theory and Genomics
新智元 · wechat · 2026-07-14
This report highlights two cases:
- Yuji Tachikawa, a mathematical physicist at the University of Tokyo, fed string theory notes that had stalled him for half a year to Claude Fable5. The AI provided a key insight to advance his problem and even invoked SymPy to write code verifying the predictions.
- Euan Ashley, a Stanford geneticist and cardiologist, had Claude interpret his own genome according to clinical standards. Taking about 30 minutes, it identified Alzheimer's risk alleles and several drug metabolism variants. The process consumed roughly 400,000 tokens, with Claude estimating the cost at around $5.
The article emphasizes that AI is evolving from a "tool" into a "colleague" or even a "scientific collaborator," demonstrating surprising intuition and reasoning skills in highly specialized fields like theoretical physics and genomics.
More from Models
- Rumor claims Kimi faked performance by serving Claude; DeepSeek new model surprises in evals — realsohamparekh · 2026-09-11
- GPT-5.6 writes well but is instantly forgettable, user complains — BasedRaddka · 2026-09-11
- Opus Refuses Protein Research Codebase Over 'Safety' Concerns, Dev Considers Rolling His Own — josephdviviano · 2026-09-11
- User Hails Unconfirmed 'DeepSeek 4.1 Flash' as an Inflection Point in LLMs — himanshustwts · 2026-09-11
- Terminal Bench v4: GLM-5.3 Leads at 41.9%, Kimi-K3 Underwhelms at 12.6% — Ok_Warning2146 · 2026-09-11
- GPT-6 Astra beats Factorio with enemies in 44 in-game hours at ~$4,500 API cost — liminal_bardo · 2026-09-11