Gary Marcus bet $100,000 that AI will still miss five general-intelligence milestones by 2029
ilkamoi · reddit · 2026-07-24
A Reddit post resurfaced a 2022 Gary Marcus essay that lays out five 2029 predictions about AI’s limits.
The five claims
- AI still will not reliably understand movies at the level of narrative, motivations, and conflicts.
- AI still will not reliably answer deep questions about novels beyond literal text.
- AI still will not be able to work as a competent cook in an arbitrary kitchen.
- AI still will not be able to reliably write more than 10,000 lines of bug-free code from natural-language specs for non-experts.
- AI still will not be able to turn natural-language mathematical proofs into symbolic forms suitable for verification.
Marcus says that if a system beats at least three of these by 2029, the argument for genuine general intelligence gets much stronger; if not, he wins the bet and mentions a $100,000 stake.
Related event: Gary Marcus Bets Heavily Against AI Surpassing Top Scientists Soon(2 posts)→
More from AGI Musings
- As models saturate human work, the RL task surface shifts to industry-transformed workflows — nirmal_dist · 2026-10-26
- You traded friends for strangers in the feed; next come AI personas — r0ck3t23 · 2026-10-26
- CS professor: don't hide your AI use from future employers, lead with it — lemire · 2026-10-26
- Engineer: AI-cheap engineering will change which manufacturing jobs get a yes — burhop · 2026-10-26
- FixItBench: the joke benchmark where robot-agent teams race to repair electron microscopes — teortaxesTex · 2026-10-26
- Instructor advice: record and save all your teaching materials for future AI use — prof_g · 2026-10-26