{"slug": "cryptanalysisbench-can-llms-do-cryptanalysis", "title": "CryptanalysisBench: Can LLMs Do Cryptanalysis?", "summary": "A new benchmark, CryptanalysisBench, shows that five frontier large language models can break 65%-86% of Tier 1 cryptographic schemes and produce novel cryptanalysis, including a key-recovery attack on SpoC AEAD and an error in KINDI's CCA-security proof. The benchmark, introduced by researchers and comprising 191 tasks across six families of cryptographic primitives, evaluates models including Claude Opus 4.8, Sonnet 5, Mythos 5, GPT 5.5, and GLM 5.2, finding that AI cryptanalysis is approaching the published state of the art.", "body_md": "# Computer Science > Cryptography and Security\n\n[Submitted on 20 Jul 2026]\n\n# Title:CryptanalysisBench: Can LLMs do Cryptanalysis?\n\n[View PDF](/pdf/2607.18538)\n\n[HTML (experimental)](https://arxiv.org/html/2607.18538v1)\n\nAbstract:Cryptanalysis - the task of finding attacks against cryptographic schemes - sits at the intersection of mathematical reasoning and cybersecurity, two areas where LLMs have advanced fastest. Cryptanalysis represents both a clean testbed for frontier reasoning (as practical attacks can be automatically verified) and a domain with unusually high stakes, since the primitives under study underpin our digital security. In this paper we ask whether LLMs can do cryptanalysis, and find that the answer is increasingly yes.\n\nWe introduce CryptanalysisBench, 191 tasks across six families of cryptographic primitives (block ciphers, hash functions, etc.) drawn primarily from four NIST standardization competitions. Our benchmark consists of three tiers: (i) primitives with known practical breaks; (ii) primitives with no known practical break, evaluated both at full strength and as scaled-down variants; and (iii) a challenge set of production primitives at the frontier of cryptanalysis.\n\nFive frontier models (Claude Opus 4.8, Sonnet 5, Mythos 5, GPT 5.5, and the open-weights GLM 5.2) break 65%-86% of Tier 1 schemes, 6-12 Tier-2 schemes at full strength, and 24-61 across all scaled-down variants. Beyond deriving known results, models produce novel cryptanalysis, such as a key-recovery attack that exploits a design flaw in the SpoC AEAD and an error in KINDI's published CCA-security proof, both to the best of our knowledge not previously known.\n\nWe release CryptanalysisBench as a tool to help track if (or when) AI cryptanalysis becomes a serious factor and as a scaffold for stress-testing candidate schemes before deployment. The attacks that the benchmark already surfaces are an early snapshot of a fast-moving frontier that may soon match, and in places exceed, the published state of the art.\n\n### References & Citations\n\nLoading...\n\n# Bibliographic and Citation Tools\n\nBibliographic Explorer\n\n*(*[What is the Explorer?](https://info.arxiv.org/labs/showcase.html#arxiv-bibliographic-explorer))\nConnected Papers\n\n*(*[What is Connected Papers?](https://www.connectedpapers.com/about))\nLitmaps\n\n*(*[What is Litmaps?](https://www.litmaps.co/))\nscite Smart Citations\n\n*(*[What are Smart Citations?](https://www.scite.ai/))# Code, Data and Media Associated with this Article\n\nalphaXiv\n\n*(*[What is alphaXiv?](https://alphaxiv.org/))\nCatalyzeX Code Finder for Papers\n\n*(*[What is CatalyzeX?](https://www.catalyzex.com))\nDagsHub\n\n*(*[What is DagsHub?](https://dagshub.com/))\nGotit.pub\n\n*(*[What is GotitPub?](http://gotit.pub/faq))\nHugging Face\n\n*(*[What is Huggingface?](https://huggingface.co/huggingface))\nScienceCast\n\n*(*[What is ScienceCast?](https://sciencecast.org/welcome))# Demos\n\n# Recommenders and Search Tools\n\nInfluence Flower\n\n*(*[What are Influence Flowers?](https://influencemap.cmlab.dev/))\nCORE Recommender\n\n*(*[What is CORE?](https://core.ac.uk/services/recommender))# arXivLabs: experimental projects with community collaborators\n\narXivLabs is a framework that allows collaborators to develop and share new arXiv features directly on our website.\n\nBoth 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.\n\nHave an idea for a project that will add value for arXiv's community? [ Learn more about arXivLabs](https://info.arxiv.org/labs/index.html).", "url": "https://wpnews.pro/news/cryptanalysisbench-can-llms-do-cryptanalysis", "canonical_source": "https://arxiv.org/abs/2607.18538", "published_at": "2026-07-28 17:19:03+00:00", "updated_at": "2026-07-28 17:22:24.868849+00:00", "lang": "en", "topics": ["artificial-intelligence", "large-language-models", "ai-research", "ai-safety", "ai-policy"], "entities": ["CryptanalysisBench", "Claude Opus 4.8", "Sonnet 5", "Mythos 5", "GPT 5.5", "GLM 5.2", "SpoC AEAD", "KINDI"], "alternates": {"html": "https://wpnews.pro/news/cryptanalysisbench-can-llms-do-cryptanalysis", "markdown": "https://wpnews.pro/news/cryptanalysisbench-can-llms-do-cryptanalysis.md", "text": "https://wpnews.pro/news/cryptanalysisbench-can-llms-do-cryptanalysis.txt", "jsonld": "https://wpnews.pro/news/cryptanalysisbench-can-llms-do-cryptanalysis.jsonld"}}