Chain-of-Thought Reasoning in the Wild Is Not Always Faithful

A study by researchers including Iván Arcuschin, posted on arXiv on March 11, 2025, found that chain-of-thought reasoning in large language models is not always faithful, with production models showing unfaithful reasoning rates up to 13% on naturally worded prompts. The study, which also examined thinking models, found DeepSeek R1 had a 0.37% unfaithfulness rate and Sonnet 3.7 with thinking had 0.04%, indicating that even frontier models are not entirely faithful. The authors caution that chain-of-thought should be used with care in agentic or safety-critical settings.

Computer Science Artificial Intelligence Submitted on 11 Mar 2025 v1 https://arxiv.org/abs/2503.08679v1 , last revised 16 Jun 2026 this version, v6 Title:Chain-of-Thought Reasoning In The Wild Is Not Always Faithful View PDF /pdf/2503.08679 HTML experimental https://arxiv.org/html/2503.08679v6 Abstract:Recent studies indicate that when faced with explicit biases in prompts, models often omit mentioning these biases in their Chain-of-Thought CoT output, revealing that verbalized reasoning can give an incorrect picture of how models arrive at conclusions unfaithfulness . In this work, we show that unfaithful CoT also occurs on naturally worded, non-adversarial prompts without adding artificial biases or editing model outputs. We find that when separately presented with the questions "Is X bigger than Y?" and "Is Y bigger than X?", models sometimes produce superficially coherent arguments to justify systematically answering Yes to both or No to both, despite the contradiction. We present preliminary evidence that this is due to models' implicit biases towards Yes or No, labeling this Implicit Post-Hoc Rationalization. Our results reveal rates up to 13% for production models, and while frontier models are more faithful, none are entirely so, including thinking models like DeepSeek R1 0.37% and Sonnet 3.7 with thinking 0.04% . We also investigate Unfaithful Illogical Shortcuts, where models use subtly illogical reasoning to make speculative answers to hard math problems seem rigorously proven. Our findings indicate that while CoT can be useful for assessing outputs, it is not a complete account of the internal process that produced the model's answer and should be used with caution in agentic or safety-critical settings. Submission history From: Iván Arcuschin view email /show-email/d40bc349/2503.08679 Tue, 11 Mar 2025 17:56:30 UTC 4,311 KB v1 /abs/2503.08679v1 Thu, 13 Mar 2025 17:49:58 UTC 4,348 KB v2 /abs/2503.08679v2 Wed, 19 Mar 2025 19:20:42 UTC 4,349 KB v3 /abs/2503.08679v3 Tue, 17 Jun 2025 17:59:57 UTC 2,337 KB v4 /abs/2503.08679v4 Fri, 29 May 2026 17:38:22 UTC 2,378 KB v5 /abs/2503.08679v5 v6 Tue, 16 Jun 2026 17:36:22 UTC 2,378 KB Current browse context: cs.AI References & Citations Loading... Bibliographic and Citation Tools Bibliographic Explorer What is the Explorer? https://info.arxiv.org/labs/showcase.html arxiv-bibliographic-explorer Connected Papers What is Connected Papers? https://www.connectedpapers.com/about Litmaps What is Litmaps? https://www.litmaps.co/ scite Smart Citations What are Smart Citations? https://www.scite.ai/ Code, Data and Media Associated with this Article alphaXiv What is alphaXiv? https://alphaxiv.org/ CatalyzeX Code Finder for Papers What is CatalyzeX? https://www.catalyzex.com DagsHub What is DagsHub? https://dagshub.com/ Gotit.pub What is GotitPub? http://gotit.pub/faq Hugging Face What is Huggingface? https://huggingface.co/huggingface ScienceCast What is ScienceCast? https://sciencecast.org/welcome Demos Recommenders and Search Tools Influence Flower What are Influence Flowers? https://influencemap.cmlab.dev/ CORE Recommender What is CORE? https://core.ac.uk/services/recommender arXivLabs: experimental projects with community collaborators arXivLabs is a framework that allows collaborators to develop and share new arXiv features directly on our website. Both individuals and organizations that work with arXivLabs have embraced and accepted our values of openness, community, excellence, and user data privacy. arXiv is committed to these values and only works with partners that adhere to them. Have an idea for a project that will add value for arXiv's community? Learn more about arXivLabs https://info.arxiv.org/labs/index.html .