Ethan Mollick: Leap in AI Model Capabilities Followed by Regression to the Mean
emollick · x · 2026-07-20
Professor Ethan Mollick highlights a peculiar phenomenon in AI: some models exhibit exceptionally strong capabilities, but their subsequent iterations often "regress to the mean," performing worse than expected.
He cites Anthropic's Updated Sonnet 3.5 (sometimes called Sonnet 3.6) as an example, suggesting it represents an extremely rare, breakthrough leap in capability relative to its predecessor.
More from Models
- Daily AI brief: GPT-Live-1 in API, OpenAI pauses $200 Pro signups amid Astra demand — koltregaskes · 2026-09-11
- Same Echo Maze prompt, three frontier models: all passed visually but shipped the same hidden bug — eyishazyer · 2026-09-11
- Benchmark scores drop from 89% to 19% on new evals — how benchmaxxing breaks leaderboard trust — airesearch12 · 2026-09-11
- ChatGPT tells user their question is too hard and to 'accept dumber answers' — phido3000 · 2026-09-11
- Developer Building a Unified Leaderboard of All Model Benchmark Scores — airesearch12 · 2026-09-11
- Rumor claims Kimi faked performance by serving Claude; DeepSeek new model surprises in evals — realsohamparekh · 2026-09-11