An AI broke an Enigma message unsolved since 2005.
A researcher pointed OpenAI's GPT-6 Astra at an 82-letter German army message from July 1941. The model built its own Enigma simulator, wrote the cracking code and tested 14.8 million keys. It found the key in about ten hours, and the historian who lists unbroken messages confirmed the result.
Why it mattersThe message asked for marching orders and named a village, not a secret worth keeping. What counts is that a general-purpose model did work a specialist team once needed.
Read in the app — free, in 9 languages
Related stories
A model trained only on ancient Greek is restoring broken papyri.2026-09-22Asked to stab a doll, the best AI model refused only twice.
2026-09-22A university lab got a 550-billion-parameter model onto a laptop.
2026-09-22OpenAI says one model resolved more than 100 open maths problems.
2026-09-22Chatbots got money questions wrong more than half the time.
2026-09-21