Berkeley workshop readout: Why public participation in AI feedback is costly
beenwrekt · x · 2026-09-16
Jessica Dai's second readout on arg min of the UC Berkeley microconference on "public feedback for AI and beyond" focuses on the burdens participants face when providing feedback.
Her premise: the public has important things to say about AI experiences but decision makers rarely listen; evaluations and aggregated information can influence consequential decisions — so public-feedback-based AI evaluations could address the misalignment between AI developers and everyone else.
This post examines the costs participants actually experience: engagement is 'costly' (why human subjects studies compensate participants), and self-selection given these costs undermines representativeness. Costs also manifest concretely in ways hard to quantify economically, and extend beyond the initial decision to participate.
More from AGI Musings
- Elad Hazan Recalls Two Decades of Online Convex Optimization — Born as a Job Hedge — HazanPrinceton · 2026-09-16
- Microsoft AI chief Suleyman calls to strip AI consciousness talk from training docs, jabbing Anthropic — pstAsiatech · 2026-09-16
- Cambridge AI-risk scholar on whistleblower warnings: >10% extinction odds this decade — S_OhEigeartaigh · 2026-09-16
- Ben Bajarin 'supremely bullish' on AI buildout after Pat Gelsinger podcast — BenBajarin · 2026-09-16
- NYT Profiles the Recursive Self-Improvement Startups, Including Jeff Clune's Recursive Superintelligence — jeffclune · 2026-09-16
- Investor pushes back on AI sentience hype: 'AI are software programs' — firstadopter · 2026-09-16