Simulated reason models simulate reasoning
2025-07-22 19:31:45.47121+02 by Dan Lyke 0 comments
New Apple study challenges whether AI models truly “reason” through problems
In early June, Apple researchers released a study suggesting that simulated reasoning (SR) models, such as OpenAI's o1 and o3, DeepSeek-R1, and Claude 3.7 Sonnet Thinking, produce outputs consistent with pattern-matching from training data when faced with novel problems requiring systematic thinking. The researchers found similar results to a recent study by the United States of <span data-huuid="4951512150826999828">America Mathematical Olympiad</span> (USAMO) in April, showing that these same models achieved low scores on novel mathematical proofs.
The article is pretty good, rounding up a bunch of pro and con arguments.