According to Woofun AI, OpenAI's GPT-6 Astra model successfully deciphered the WWII Enigma code, causing the market to misinterpret Bitcoin's security. Carter Levin used this model to process historical ciphertext to prove the computing power advantage of AI in specific scenarios, but this does not pose a threat to the underlying logic of modern cryptocurrencies.
On July 10, 1941, a German soldier sent an 82-character radio message with directions to the town in Rozenau. This information lay dormant in the CryptoCellar public archives for 83 years and remains uninterpreted. Carter Levin of Bloomberg's product development department in New York handed over the task to the GPT-6 Astra model, and it only took 10 hours to obtain plain text. By comparing the word 'ROSENOW' in the deciphered message on the same day, Astra speculated that the same term existed in the undeciphered text. Afterwards, it independently built an Enigma password simulator, wrote the attack code, and ran various key combinations to compare it with the Telegram header information.
According to data compiled by Woofun AI, the model tested a total of about 14.8 million sets of keys and finally locked down the answers. The authenticity is supported by details: the recovered German message preserves the sender's mistake of misspelling 'BITTE' as' BTTE ', a state of confusion that is more credible than a fake clean result.
Enigma ciphers are hardware encryption from the 1930s, and their security was drastically reduced after having reminders, so a normal laptop can handle it. In contrast, Bitcoin uses a 256-bit key based on the secp256k1 curve, which requires about 2^128 operations to crack, which exceeds the total number of atoms visible in the sky. Even if you put in 10 hours of computing power, even extending it to 10 billion years, you won't be able to have a real impact on it. There is an essential difference in cryptographic strength between the two, and AI breakthroughs have not shaken Bitcoin's mathematical roots.
The real risk does not come from algorithm cracking, but rather from implementation level bugs. AI is good at solving search problems, not brute force cracking. In October 2025, about 220,000 wallets had design flaws due to random number generator flaws, and this hidden danger was ignored for a long time.
Additionally, brainwave wallets, forgotten mnemonics, and unaudited contracts are all high-risk factors. As long as a reliable wallet is used and the key is generated correctly, the elliptic curve encryption technology is still stable, and users should focus on their own operational safety rather than excessive panic.