David Khourshid: curiosity, pushback and rest are the human edges agents can't fake
DavidKPiano · x · 2026-09-10
David Khourshid argues the human-agent gap in development isn't capability but three things that look like weaknesses: curiosity (humans obsess over unrequested ideas; agents only explore inside someone else's prompt), pushback (human skepticism comes from scar tissue and maintenance stakes; agent pushback is simulated obedience), and rest (stopping bounds the blast radius and catches debt; overnight agent loops accumulate cognitive debt with unshrinking state spaces, and degraded judgment never 'feels' degraded).
All three reduce to one question — what is worth doing — which only matters to a persistent self that pays for its own mistakes. He argues no prompt, skill, or harness genuinely fixes this, and it won't be solved in the next model update.
More from AGI Musings
- AI execs' common refrain: things are scary but we're trapped in a prisoner's dilemma — SatelliteNetSec · 2026-09-11
- Richard Socher: goal setting is the key to controlling superintelligent AI — RichardSocher · 2026-09-11
- Defending AI risk concerns: the marketing-conspiracy theories don't hold up — AndyMasley · 2026-09-11
- AI to cure all diseases in 5 years then end humanity, quip sparks mockery — MaxUnfried · 2026-09-11
- Is quantifying uncertainty arbitrary? A Bayesian ideology debate — NathanpmYoung · 2026-09-11
- Zvi disputes claim that AI lost a live debate against conservatives in Dallas — TheZvi · 2026-09-11