Jev and the bittersweet lesson: can differently sized models serve different apps?
amankhan · x · 2026-09-21
Riffing on a week dominated by Jev paradox discourse, the author explores what it teaches about the Bitter Lesson: can models of different shapes and sizes combine into systems predicting beyond text, images, or numbers? Can models forecast action probabilities? Are we still waiting for a general-purpose time-series regression model, new modalities, or paradigm improvements driven by emerging tasks?
More from AGI Musings
- Why people hate AI: a long-simmering list of grievances finally erupts — TuhinChakr · 2026-09-21
- Richard Sutton flags a proposal to rescue Grokipedia via an open two-sided market for scientific data — RichardSSutton · 2026-09-21
- Drawing out the inner/outer loop of judgement: autonomy vs accountability — _AustinCalvert_ · 2026-09-21
- When AI proves theorems faster than mathematicians can absorb them, what happens to math? — burny_tech · 2026-09-21
- Don't Stop AI — Build It Responsibly: Move Fast, But Measure Faster — AryHHAry · 2026-09-21
- Debate: You Can't 'Premove' Against Superintelligence, Precautions Must Come Before It Arrives — ctjlewis · 2026-09-21