Reddit debate: 'fancy autocomplete' vs 'imminent RSI' camps, and one user's conversion
Mathufnn · reddit · 2026-11-03
A Reddit long-post maps the polarized AI discourse: one camp sees models as fancy autocomplete inflated by marketing; the other believes RSI/AGI is 1-2 years away, citing autonomous coding and frontier-research capabilities. The author, once a skeptic, says Claude Opus 5.5 converted them to expecting AI to soon outperform humans on virtually all computer tasks — while admitting they have zero evidence beyond personal experience.
More from AGI Musings
- Anthropic warns in S-1 filing it's very close to recursive self-improvement capability — emmanuelvivier · 2026-11-04
- Economist: AI already persuades better than 99% of humans — 'we will be its agents' — paulnovosad · 2026-11-04
- Econ KOL warns AI is already smarter than 99% of humans and will make us its agents — paulnovosad · 2026-11-04
- GI Bill history: the 'idleness panic' of 1944 may be wrong about ASI too — krishnan · 2026-11-04
- Telocracy: A Vision Where AI Holds Institutions Accountable to Their Purposes — sebkrier · 2026-11-04
- AI Made Creating Cheap — Taste Is the Skill That Matters Now — aitrendz_xyz · 2026-11-04