I’ve been working on a hobby project to crack classical ciphers. The guts of it is to see whether frontier LLMs, given the right tools, can do a good job cracking a broad range of historical ciphers. https://github.com/matthewdgreen/decipher

I’ve been working on a hobby project to crack classical ciphers. The guts of it is to see whether frontier LLMs, given the right tools, can do a good job cracking a broad range of historical ciphers. https://github.com/matthewdgreen/decipher

What’s interesting is how slow and challenging this has been. Frontier LLMs “know” a lot about cipher cracking, but they really struggle to operationalize it. There’s a lot of creativity in it. Without tools, even top-tier coding agents take forever and give up.