Skip to content
worth noting Security

Cryptographer Matthew Green comments on AI cryptanalysis work by Anthropic

only one source so far

Cryptographer Matthew Green responded to recent cryptographic work by Anthropic and pointed out that the emergence of AI capable of cryptanalysis comes at the very time of the transition to post-quantum cryptographic algorithms.

Simon Willison quoted a comment on his blog by cryptographer Matthew Green on recent work by Anthropic in cryptography. Green points out that we are now in the midst of a transition from traditional public cryptographic algorithms based on elliptic curves and RSA to new post-quantum algorithms addressing new mathematical problems — which is precisely why a number of new standards, such as HAWK, are being considered today.

According to Green, this is either the ideal moment or, conversely, the worst possible moment for a major new AI cryptanalysis capability to emerge. If AI does not succeed in breaking all the hard mathematical problems used to date (or if we do not live in the so-called Minicrypt of Impagliazzo), then, in his view, right now is the best time for AI to improve at cryptanalysis. In an optimistic scenario, Green says, this will lead to real confidence in the problems underpinning post-quantum algorithms, and the cryptanalysis literature will become much more robust.

The source does not specify the content of the cryptographic work by Anthropic that Green is commenting on. You can find details in the source article.

What changed

Why it matters

The comment shows that the expert community is discussing the two possible effects of using AI for cryptanalysis at the sensitive moment of the transition to post-quantum standards: it could either help verify the resilience of newly proposed algorithms before they become widely used in practice, or reveal weaknesses in the assumptions underpinning current secure communication. However, the source does not mention specific results or a product; this is a general reflection on the risk and opportunity.

AI cryptanalysis Anthropic cryptography Matthew Green post-quantum cryptography security

Check the original

Event sources

only one source so far · 1 publisher, 1 independent. We count feeds from the same owner only once.

1
Simon Willison — AI tag (leading independent LLM commentator) community signal · first detected Quoting Matthew Green