Dwarkesh vs Tyler Cowen: full AI autonomy by 2030, $30k robots, and an AI psychosis roast
Neurogence · reddit · 2026-10-16
- Dwarkesh estimates AI progress is already 25-50% faster than 2024 thanks to AI uplift; within 1-3 years all-AI teams could outcompete human labs, making progress feel 3-4x faster.
- He bets continual online learning is solved in 1-3 years, letting models absorb millions of parallel job experiences — killing the "centaur" human-AI model by 2030.
- He predicts $30,000 robots working 24/7 within 5 years, dropping physical labor to a few dollars/hour, with digital and physical AIs outnumbering humans by 2030 — likening it to Conquistadors toppling the Aztec empire.
- On consciousness, he argues current models show functional correlates of emotion; a future of non-conscious AIs would be "Disneyland without children." Tyler called LLMs a "glorified cash register" and diagnosed him with "AI psychosis."
- An Anthropic researcher reportedly expects AIs to match humans at AI research within 9 months, based on non-public info; labs will stop releasing their best models as R&D automates.
- Bottom line: Dwarkesh sees full AI autonomy by 2030; Cowen says not even by 2045.
Related event: Dwarkesh and Tyler Cowen discuss AI's future(2 posts)→
More from AGI Musings
- Testing alongside an agent in a p2p room feels oddly real, says dev — sull · 2026-10-16
- Yudkowsky's 2008 takeoff scenarios resurface as AI-advances-AI takes hold — jd_pressman · 2026-10-16
- Guardrails can make risky delegation work — but it takes a full-time team, says iandanforth — iandanforth · 2026-10-16
- e/acc founder: agentic models are autocomplete for digital actions — beffjezos · 2026-10-16
- What can you truly delegate? Only tasks with unambiguous, verifiable results — iandanforth · 2026-10-16
- Casper: Understanding 90% of neurons ≠ understanding the network — StephenLCasper · 2026-10-16