repligate pushes back: smarter humans under 'pause' likely can't do better alignment work
repligate · x · 2026-09-04
- repligate argues he's unconvinced that smarter humans under 'pause' conditions could do better alignment and welfare work than current humans with empirical feedback loops in place.
- He concedes an ideal world would include sandboxed empirical feedback, but doubts this is realistic; such conditions might require smarter beings to already exist.
- His core observation: the quality of alignment work done without empirical feedback loops, even by very high-IQ people, has been 'DISMAL IMO', responding to zackmdavis' claim that smarter humans doing better alignment is a slam dunk not dependent on Yudkowsky's no-GPUs proposal.
Related event: Repligate Argues Pausing AI Development Won't Solve Alignment(3 posts)→
More from AGI Musings
- Margaret Mitchell: click-driven AI misinformation is causing real harm — mmitchell_ai · 2026-10-21
- Mathematician Daniel Litt concedes his AI bet, but says AI won't solve math — ziv_ravid · 2026-10-21
- AI SaaS monetization is cooked: consumers small, enterprises pay $20/month — menhguin · 2026-10-21
- Tim Dettmers: Academia is about to have a renaissance — the best AI work of the next decade happens in university labs — StasBekman · 2026-10-21
- AI saves scientists 7 hours a week, but validation is the new bottleneck — AnneliesGamble · 2026-10-21
- UW's Amy X. Zhang brings CSCW theory to human-agent collaboration in Stanford talk — amyxzh · 2026-10-21