Using LLMs less fixed his stuck project: on keeping AI out of your thinking
yacineMTB · x · 2026-10-09
internetvin (retweeted by yacineMTB) shares a practical reflection: after spinning in circles on a project, he used LLMs less — treating them as a sophisticated search engine and removing them from his thinking process — and quickly figured out the problem.
His core point: it's far easier than most realize to give LLMs too much weight. Without discipline about developing and tracking your own understanding, it's like having someone constantly commenting on every idea — pushing you down unrelated paths or shutting down promising exploration earlier than it should.
More from AGI Musings
- SubDGuide: modeler-inspired agentic workflow for mesh-to-SubD reconstruction — ssh4net · 2026-10-10
- MIT paper: AGI's binding constraint is human verification bandwidth, not intelligence — Afinetheorem · 2026-10-10
- Jay Kreps predicts humans won't be needed for math in 3 years; peers push back — JosephJacks_ · 2026-10-10
- "Solving it kills the grants": AI in math rekindles academia critique — _AustinCalvert_ · 2026-10-10
- "Math belongs to taxpayers": AI solving math sparks ownership debate — _AustinCalvert_ · 2026-10-10
- OpenAI's Noam Brown Delivers Concluding Remarks on AI Alignment at COLM 2026 — erichorvitz · 2026-10-10