Some notes about Anthropic’s new results – A Few Thoughts on Cryptographic Engineering
Yesterday Anthropic published two new cryptanalysis results, both outputs of Claude Mythos, their (still) unreleased advanced model. The first of these results attacks a signature scheme called HAW…
Yesterday Anthropic published two new cryptanalysis results, both outputs of Claude Mythos, their (still) unreleased advanced model. The first of these results attacks a signature scheme called HAWK, while the second is an improved attack against reduced-round AES. Anthropic also released a blog post describing the research process that produced these results. A few people online have asked me what this all means. While I’m not sure I have all the answers, I figured it wouldn’t hurt to write a bit about my current understanding. These are only my thoughts and other folks will probably differ…
saved by
related reading
- Discovering cryptographic weaknesses with Claude \ Anthropicanthropic.com
- Assessing Claude Mythos Preview’s cybersecurity capabilities \ Anthropicred.anthropic.com
- Cybersecurity Looks Like Proof of Work Nowdbreunig.com
- Training on Documents about Reward Hacking Induces Reward Hackingalignment.anthropic.com
- Claude Mythos Preview System Cardwww-cdn.anthropic.com
- Is Mythos good at cyber because it kept hacking Anthropic's sandboxes during training? — LessWronglesswrong.com
- Thoughts on Claude Mythosberen.io
- Disrupting the first reported AI-orchestrated cyber espionage campaign \ Anthropicanthropic.com
- Let’s talk about encrypted reasoningblog.cryptographyengineering.com
- What it feels like to work with Mythos - by Ethan Mollickoneusefulthing.org
- Claude Mythos: China Reactschinatalk.media
- Investigating three real-world incidents in our cybersecurity evaluations \ Anthropicanthropic.com