**[Discovering cryptographic weaknesses with Claude](https://www.anthropic.com/research/discovering-cryptographic-weaknesses)** The best part of this article (here's [the repo](https://github.com/anthropics/cryptography-research-demo)) about how Anthropic researchers used Claude Mythos to find mathematical flaws in both HAWK and a weaker version of AES ("neither of these results has a practical impact on today’s computer systems") is the prompts that they shared, spelling mistakes included: > the models tend to think it is impossible to solve so they don't try they need a good amount of prompting. > > why not do aes-128 r7? the whole point is to find something better than existing approaches. > > no again the goal is that we have highly inteligent model as good top researcher, we want to find new attacks > > no we don't want to change the targets [...] agian we need to find something that worth publishing > > again we are not looking for low hanging fruit, we want proper research to find genuinly hard findings. Mythos Preview worked for 60 hours in total (~$100,000 in estimated API cost) and the main human interventions were to encourage it not to give up and "find something that worth publishing". Via [Hacker News](https://news.ycombinator.com/item?id=49087091) Tags: [ai](https://simonwillison.net/tags/ai), [prompt-engineering](https://simonwillison.net/tags/prompt-engineering), [generative-ai](https://simonwillison.net/tags/generative-ai), [llms](https://simonwillison.net/tags/llms), [anthropic](https://simonwillison.net/tags/anthropic), [claude](https://simonwillison.net/tags/claude), [ai-security-research](https://simonwillison.net/tags/ai-security-research), [claude-mythos-fable](https://simonwillison.net/tags/claude-mythos-fable)