Richard Sutton Says Brute-Force Compute Won't Yield Intelligence, Sparking Fresh Bitter Lesson Debate

Turing Award winner and reinforcement learning pioneer Richard Sutton posted a brief statement on November 8: brute force computation alone is insufficient for intelligence—we still need clever new methods, and the science of AI is far from finished. The statement was widely read as a direct reflection on, and pushback against, his own "Bitter Lesson" thesis and the current compute-scaling路线.

Confirmed

• Sutton explicitly stated that brute-force computation cannot produce intelligence, that clever new methods are needed, and that AI science is far from complete.
• Several figures in the field quickly responded publicly, sparking an open debate.

Reactions

• John Langford argued that the cycle of "scale, invent, repeat" is far older than the "Bitter Lesson" debate, and dug up a 2011 episode: back then some claimed language modeling was solved by n-gram methods—implying that "scaling is over, we need new methods" talking points have recurred throughout history.
• Former OpenAI researcher Aroohan commented that this would frustrate many young scientists who, believing in the "Bitter Lesson" and that scaling alone suffices, were misled.
• Well-known researcher giffmana quote-posted to mock the view, a direct rebuttal to Sutton.
• Investor Rohit Krishnan reshared it as "Bitter bitter lesson," implicitly questioning Sutton's stance.

Why it matters

• Sutton, author of the original "Bitter Lesson" essay and a Turing Award winner in reinforcement learning, turning against his own thesis is itself a landmark moment, touching the field's most central debate: does intelligence come mainly from stacking compute and scale, or from new scientific and methodological breakthroughs?
• Both sides feature prominent figures (Langford's historical perspective, rebuttals from giffmana and Krishnan, Aroohan's concern for young researchers), reflecting a split in confidence in the scaling path across academia and investment circles.

2026-11-08 ~ 2026-11-08 · 10 related posts

• [source] Richard Sutton: Brute force compute isn't enough, the science of AI is far from finished — RichardSSutton · 2026-11-08
• Sutton says brute-force compute won't yield intelligence; researcher laments scaling's toll on young scientists — eigenron · 2026-11-08
• Sutton: brute force isn't intelligence; Langford cites 2011 n-gram era to frame scale-invent cycle — weijie444 · 2026-11-08
• Rich Sutton walks back the Bitter Lesson: brute force isn't enough for intelligence — eigenron · 2026-11-08
• Sutton restates the Bitter Lesson; Dileep George fires back: 'don't fall for it' — anshulkundaje · 2026-11-08
• [source] Chollet: Scaling Is Only a Multiplier, Smarter Methods Scale Better — fchollet · 2026-11-08
• Richard Sutton pushes back on the Bitter Lesson: brute force isn't enough for intelligence — dbreunig · 2026-11-08
• [source] Richard Sutton: brute-force compute isn't enough, AI science far from finished — dair_ai · 2026-11-08

2 near-duplicate retellings: davidmanheim · PMinervini