Optimizing a 370-year-old cipher

Putting all controversy aside - this is a refreshing use of a model. Models are going to be very, very interesting... Reminder to wait a couple of days. I suppose I hoped enduring the pain of waiting for the day when the remainder of the Maya codices are solved by AI.

What did they say around 2020? If you don't like the content on X enough to make an unbreakable cipher anymore? Putting all controversy aside - this is a refreshing use of a model. Models are going to be a lot of learnable discrete components, or even bits and gates finally.