[2510.09714] All Code, No Thought: Current Language Models Struggle to Reason in Ciphered Language

Abstract:Detecting harmful AI actions is important as AI agents gain adoption. Chain-of-thought (CoT) monitoring is one method widely used to detect adversarial attacks and AI misalignment. However, attackers and misaligned models might evade CoT monitoring through ciphered reasoning: reasoning hidden in encrypted, translated, or compressed text. To assess this risk, we test whether models can perform ciphered reasoning. For each of 28 different ciphers, we fine-tune and prompt up to 10 models to reason in that cipher. We measure model accuracy on math problems as a proxy for reasoning ability. Across the models we test, we find an asymmetry: model accuracy can drop significantly when reasoning in ciphered text, even though models demonstrate comprehension of ciphered text by being able to translate it accurately to English. Even frontier models struggle with lesser-known ciphers, although they can reason accurately in well-known ciphers like rot13. We show that ciphered reasoning capability correlates with cipher prevalence in pretraining data. We also identify scaling laws showing that ciphered reasoning capability improves slowly with additional fine-tuning data. Our work suggests that evading CoT monitoring using ciphered reasoning may be an ineffective tactic for current models and offers guidance on constraining the development of this capability in future frontier models.

Preprint. Under review. A LL C ODE , N O T HOUGHT: C URRENT L ANGUAGE M ODELS S TRUGGLE TO R EASON IN C IPHERED L AN - GUAGE Shiyuan Guo*1 , Henry Sleight2 , Fabien Roger*3 1 Anthropic Fellows Program 2 Constellation 3 Anthropic A BSTRACT…

saved by

• Evie Hu

related reading

• [2602.23163] A Decision-Theoretic Formalisation of Steganography With Applications to LLM Monitoringarxiv.org
• [2512.11949] Neural Chameleons: Language Models Can Learn to Hide Their Thoughts from Unseen Activation Monitorsarxiv.org
• Training Large Language Models to Reason in a Continuous Latent Spacearxiv.org
• [2606.04071] Covert Influence Between Language Modelsarxiv.org
• [2502.17424] Emergent Misalignment: Narrow finetuning can produce broadly misaligned LLMsarxiv.org
• [2507.11473] Chain of Thought Monitorability: A New and Fragile Opportunity for AI Safetyarxiv.org
• [2201.11903] Chain of Thought Prompting Elicits Reasoning in Large Language Modelsarxiv.org
• [2412.06769] Training Large Language Models to Reason in a Continuous Latent Spacearxiv.org
• [2608.09867] Stealing Reasoning Traces from Proprietary LLM APIsarxiv.org
• When Chain of Thought is Necessary, Language Models Struggle to Evade Monitorsarxiv.org
• Stolen Thoughtsstolen-thoughts.com
• [2603.07267] How to Steal Reasoning Without Reasoning Tracesarxiv.org