Open-source models by 2029 could make bioweapon creation "almost trivial", debaters clash on biorisk
JMannhart · x · 2026-09-11
JMannhart and @tszzl clash over biorisk. JMannhart argues that by current trends, access to an open-source model from 2029-2030 plus a synthesis route could make biological attacks "possibly almost trivial".
@tszzl pushes back that he's "vastly underestimating biorisk": doing it accidentally (e.g. with evolution as a dumb optimization algo) is hard, but anyone half-skilled at using bio as an optimization process could manage it.
Related event: Debate Erupts Over AI Biosecurity Risk by 2029(2 posts)→
More from AGI Musings
- Tyler Cowen: name the market prices that would confirm your AI doom predictions — that's taking it seriously — zetalyrae · 2026-09-11
- When tech fluency is universal, the key skills will be non-technical — round · 2026-09-11
- Business Insider Asked ChatGPT, Gemini, Claude and Grok How AI Could End Humanity — coinfanking · 2026-09-11
- New Essay Argues AI Superintelligence May Never Be Achievable — horaciogarza · 2026-09-11
- Want to stop ASI? Quitting, laws, and treaties all fail—unless you conquer the world first — bennash · 2026-09-11
- Caltech undergrads defend AI math contest Mathathon against open letter criticism — Singularitarian · 2026-09-11