analiza aiqbench
Anthropic's Claude Mythos Preview has demonstrated that frontier AI models can now identify mathematical weaknesses in cryptographic algorithms like HAWK and reduced-round AES at speeds far exceeding traditional research methods. While these findings do not compromise currently deployed encryption systems, they signal a paradigm shift where advanced LLMs serve as potent tools for pre-deployment cryptanalysis rather than just software bug detection. This development underscores the critical need to integrate AI-driven testing into future cryptographic standardization processes to ensure robustness against algorithmic vulnerabilities before widespread adoption.
AI Finds New Weaknesses in Cryptographic Algorithms, Anthropic Says
Insider Brief Anthropic says one of its newest AI models discovered improved mathematical attacks against two important cryptographic algorithms, demonstrating that frontier AI systems may soon become valuable tools for testing the security of the encryption methods that protect digital communications. The company said the findings do not threaten today’s deployed encryption systems. Instead, they […]
Anthropic says its Claude Mythos Preview AI model discovered improved mathematical attacks against the HAWK post-quantum digital signature candidate and a reduced-round version of AES, demonstrating that advanced AI can contribute to cryptography research without affecting current production systems. The AI reduced HAWK’s effective security margin by identifying a previously unknown mathematical symmetry and developed a new attack that is 200 to 800 times faster against a research version of seven-round AES, while researchers spent weeks validating the model’s findings. Anthropic says the results highlight AI’s growing role in evaluating encryption algorithms before deployment, prompting the company to release new research papers and CryptanalysisBench to help researchers measure AI capabilities in cryptanalysis. Anthropic says one of its newest AI models discovered improved mathematical attacks…
Extras afișat conform termenilor RSS ai publicației · 124 cuvinte · textul integral și figurile sunt pe site-ul publicației
Subiecte conexe
AI cryptanalysis compresses flaw-finding time from months to hours.
Quantum Zeitgeist · 2 oct. 2026
Claude AI Solves 25-Year NSA Cryptography Mystery in Just 30 Minutes
Cybersecurity News · 17 oct. 2026
Lattice-Based Cryptography Explained
Quantum Zeitgeist · 15 aug. 2026
The J-Space Debate, Agent Swarms, and Pacing Frontier AI - Digital Minds Newsletter #4
LessWrong · 18 sept. 2026