Anthropic researchers have discovered two cryptographic attacks using the Claude artificial intelligence model. The attacks, and indeed the very possibility of their success, reveal real mathematical weaknesses in encryption algorithms.
The first method identifies a previously undetected symmetry in HAWK, a post-quantum digital signature procedure currently being tested by NIST (National Institute of Standards and Technology, USA). The discover…
This story is only covered by news sources that have yet to be evaluated by the independent media monitoring agencies we use to assess the quality and reliability of news outlets on our platform. Learn more here.
Anthropic researchers have discovered two cryptographic attacks using the Claude artificial intelligence model. The attacks, and indeed the very possibility of their success, reveal real mathematical weaknesses in encryption algorithms.
The first method identifies a previously undetected symmetry in HAWK, a post-quantum digital signature procedure currently being tested by NIST (National Institute of Standards and Technology, USA). The discover…