Quoting Terence Tao
- ID
- 22593
- Status
- summarized
- Published
- 09 Sep 2026, 8:20 AM
- Fetched
- 09 Sep 2026, 9:09 AM
- Provider
- Simon Willison
- Category
- developer-ai
- Original URL
- https://simonwillison.net/2026/Sep/9/terence-tao/
- Source URL
- https://simonwillison.net/atom/everything/
Summary
- Score
- 5.5
- Created
- 09 Sep 2026, 9:11 AM
- Tags
- Audience
- ai_ml_learnerssaas_foundersdevelopers
What happened
Terence Tao warns that AI is mining open research problems in a non-renewable way—mere rumor of someone working on a problem can trigger massive AI-powered effort to solve it before the original project matures. He argues this may push researchers to stop sharing promising directions, reversing centuries of open science tradition and damaging the field long-term.
Why it matters
If you build in open source or open research-adjacent spaces, this signals a shift in how knowledge sharing works under AI acceleration. The dynamic Tao describes—where sharing a promising direction invites immediate automated competition—could apply to open source projects, benchmarks, and technical blog posts too. Consider whether your team's approach to sharing early-stage work or roadmaps needs adjustment in an environment where AI agents can replicate and flatten ideas overnight.
Discussion angle
Does the same 'flatten on rumor' dynamic Tao describes for math research already apply to open source and indie SaaS—where publishing a novel approach or feature invites immediate AI-assisted cloning? How should builders adapt their sharing and shipping strategy?