In 2026, GPT 6 Astra cracked a 20-year-old WWII Enigma code in two days, showcasing AI's new autonomous research powers.
AI solves WWII encryption puzzle
In September 2026, the GPT 6 Astra model from Open AI independently solved a historical enigma that had remained unsolved for more than twenty years. The puzzle involved breaking a message encrypted by the famous Enigma machine used during World War II.Message decryption process
The task required analyzing a message encrypted using the Enigma machine, a device used by Nazi Germany to secure military communications. While human experts previously spent weeks reviewing archives, the AI accomplished the feat in two days.AI's methods and tools
The AI cross-referenced historical sources, identified an original typing error, and developed its own simulation tools in Python and C++. These tools allowed it to test different decryption methods and verify the accuracy of its findings.Significance of the achievement
This achievement marks the emergence of autonomous research agents capable of exploring complex databases, formulating hypotheses, and executing their own code to solve problems without human intervention. The AI demonstrated an ability to think critically and creatively, similar to human researchers.Implications for enterprises
The event highlights new capabilities for enterprises, particularly for CIOs and innovation departments. These autonomous analysis and problem-solving capabilities could soon apply to areas such as supply chain data, information security audits, and the optimization of complex processes.