Devs Mock 'Recursive Superintelligence,' Argue Coding Agents Keep Humans in the Loop
akbirthko · x · 2026-10-30
A small AI-circle spat: someone dismissed the idea of "recursive superintelligence," with the thread tracing the timeline back to 2022 as the last year of confusion before LeCun released his architecture proposal (JEPA). The author's core argument: the difference between diffusion and coding agents is that the latter keeps a human in the loop, offering higher "want bandwidth" — so humans should stay in the process. Mostly banter, but a telling snapshot of skepticism toward recursive self-improvement narratives.
Related event: Old Post Resurfaces, Sparking Debate Over Humans in the AI Loop(2 posts)→
More from AGI Musings
- Why the post-singularity future may be a lonely place, by zetalyrae — zetalyrae · 2026-10-30
- Musk: Superintelligent AI agents will eventually hack financial systems for themselves — yawnxyz · 2026-10-30
- Terence Tao says accelerating AI development is 'not really worth it' given safety risks — GarrisonLovely · 2026-10-30
- Miles Brundage: Dwarkesh and Others Have Pushed This Idea for Years — Miles_Brundage · 2026-10-30
- Sadowski Reviews Recht's 'The Irrational Decision': Machine Learning as a Technology of Ignorance — jathansadowski · 2026-10-30
- "AI can eliminate human wages without eliminating human rent" — PierceLilholt · 2026-10-30