A new study on LLMs by Apple: "We found no evidence of formal reasoning in language models," The behavior of LLMS "is better explained by sophisticated pattern matching" which the study found to be "so fragile, in fact, that [simply] changing names can alter results." arxiv.org/pdf/2410.05229

Highly suggest reading it. Definitely a great take on why LLMs are a *predictive* model, not a *reasoning* model.