A user named X claims Claude recovered a Bitcoin wallet containing 5 BTC that had been locked for 11 years, attracting 6 million XEM — but experts are questioning the true nature of the process.
A chain of posts on X by an anonymous user “Cprkrn” has garnered over 6 million XEM after the user claimed Anthropic ’s Claude AI helped recover a Bitcoin wallet worth approximately $400,000, equivalent to 5 BTC, that had been locked since 2015. The wave of Chia spread rapidly on Wednesday, sparking a wave of debate about the true limits of the analytical capabilities of large-scale linguistic models.
According to Cprkrn's description, this person repeatedly failed to use specialized tools like btcrecover and Hashcat to try password combinations before uploading data from an old college computer to Claude.
Here, the AI reportedly identified an old wallet file that the Mnemonics phrase found in a notebook successfully deciphered. “Locked out for over 11 years just because I got high and changed the password,” Cprkrn wrote. Blockchain data confirms the wallet address, starting with “14VJyS,” has had no transactions since 2015.
Digital forensic analysis, not cryptography.
However, experts recovering the wallet quickly drew a crucial line. There was no evidence in the Chia screenshots to suggest that Claude had bypassed Bitcoin's underlying cryptographic mechanisms. One expert told Decrypt that Claude's most likely Vai was sorting through large amounts of historical data and identifying clues related to old login credentials—a form of digital forensic screening, not a cryptographic attack.
In other words, what's truly unlocked is the Mnemonics phrase the user Capital possesses; Claude simply helps locate the file containing it within the chaotic data.
On Reddit, the skeptical reaction was even more intense. Many users argued that the viral headline was deliberately vague to exaggerate the Vai of AI, while the essence of the matter was much simpler: Claude acted as an intelligent search engine within a personal file repository. This event reflects a familiar trend: public expectations of AI often far exceed technical reality, and viral stories tend to exploit that gap.
The story emerged at a time when interest in Claude was heightened after Anthropic launched its new model with enhanced security analysis capabilities, but it also served as a reminder that there is a huge technical gap between "AI assisting in data investigation" and "AI cracking passwords," even though the two phrases look similar in the headline.
