Ilya Suggests the Brain May Use Backprop; Critic Calls Metaphor Off by Five Orders of Magnitude
JosephJacks_ · x · 2026-10-12
Ilya Sutskever resurfaced his argument that backprop's dominance over other neural network methods hints the brain might use backprop too.
JosephJacks pushed back: six years later there is still zero evidence the brain uses backprop, and treating synapses as "weights" is a flawed metaphor off by roughly five orders of magnitude from neural fundamentals. The exchange captures a core debate on whether biological learning resembles deep learning.
More from Research
- Using AI agents to reverse engineer and patch games with byte-exact verification — arpit_bhayani · 2026-10-12
- OpenAI matrix multiplication papers spark debate: middle-school combinatorics beat CW-style bounds — basedjensen · 2026-10-12
- Is RL generalization a myth? Skeptic argues AGI via RL needs a bespoke environment for every task — akbirthko · 2026-10-12
- Automated peer review at $0.10 per paper: dev says $500M could audit all 100M scientific papers — clarejtbirch · 2026-10-12
- RLHF sawed off 'Decline': why LLMs hallucinate APIs and the one-line Flores fix — RandalSchwartz · 2026-10-12
- Hospitals Protect the Image but Leak the Report: Preprint Benchmarks Pseudonymisation with 15 Detectors, 4 Languages — maier_ak · 2026-10-12