Researchers clash over claim that data-driven "ML for Science" is mostly nonsense
tak3sh8 · x · 2026-11-16
A contested take is making the rounds in the AI-for-science debate: the quoted post argues that "ML for Science" — data-driven approaches like PINNs and surrogate models — is mostly nonsense, while "AI for Science" — using reasoning models and formal methods to accelerate discovery — is extremely promising. The reposter mockingly objects to the blanket dismissal of data-driven methods, highlighting an ongoing rift between reasoning-based and data-driven approaches to AI-accelerated science.
More from AGI Musings
- Blogger: the ultimate breakthrough is manufacturing ASI until intelligence is no longer scarce — Dr_Singularity · 2026-11-16
- Robin Hanson: Doomsday theories must predict smaller testable events — QuintinPope5 · 2026-11-16
- US CS Enrollment Falls 12%+ This Fall, Down 18% Over Two Years as AI Hits Junior Jobs — Fowe · 2026-11-16
- Writer using AI to rework drafts gets flagged as AI 100% of the time — aronchick · 2026-11-16
- AI takeover creating conscious observers could imply long timelines, not doom — jessi_cata · 2026-11-16
- "Ads are prompt injections for humans" — a developer's analogy that reframes both fields — maksym_andr · 2026-11-16