Chain-of-Thought doesn't make LLMs smarter; it just makes them 'talk' more while they double down on their own biases.
April 14, 2026
Original Paper
Thinking Fast, Thinking Wrong: Intuitiveness Modulates LLM Counterfactual Reasoning in Policy Evaluation
arXiv · 2604.10511
The Takeaway
The study shows that reasoning steps only help when the answer is intuitive. When faced with counter-intuitive logic, models use the extra compute to justify their initial wrong intuition, proving that LLM reasoning is often just a performative mimicry of thought.
From the abstract
Large language models (LLMs) are increasingly used for causal and counterfactual reasoning, yet their reliability in real-world policy evaluation remains underexplored. We construct a benchmark of 40 empirical policy evaluation cases drawn from economics and social science, each grounded in peer-reviewed evidence and classified by intuitiveness -- whether the empirical finding aligns with (obvious), is unclear relative to (ambiguous), or contradicts (counter-intuitive) common prior expectations.