Drake warned that the mathematics underpinning Bitcoin and Ether wallet signatures—specifically elliptic curves—follow predictable patterns that a sufficiently powerful AI could learn to exploit. By contrast, hash functions, which convert data into fixed-length digital fingerprints, are designed to minimize pattern exposure and resist reverse-engineering.
AI-assisted attacks on cryptocurrency infrastructure have already resulted in measurable financial losses.
In December, Anthropic researchers demonstrated that frontier AI models could generate working exploits against simulated versions of real DeFi contracts. In late July, a volunteer initiative known as the Bitcoin Red Team deployed AI models to scan 390 Bitcoin software projects in roughly 27 hours, identifying nearly 5,000 potential flaws, 85 of which were classified as critical.
On July 30, an attacker began draining Coldcard hardware wallets by exploiting a five-year-old firmware bug, absconding with at least 1,367 BTC. Maker Coinkite stated it suspected AI involvement in locating the vulnerability.
Days later, BTCPay Server confirmed that attackers had siphoned funds from merchants’ Lightning nodes via a flaw first uncovered during an AI-assisted audit. On August 27, Core Lightning developers issued an emergency advisory after AI-generated bug reports revealed genuine vulnerabilities in their software.
Separately, researchers used AI coding agents to refine a component of a prospective quantum attack, as CoinDesk reported in September, though the work still required quantum hardware and addressed only part of the overall attack vector.
Also Read
- Yen Gains Against Euro Amid Global Currency Shifts
- THORChain recorded $3 million September income, with 63 % driven by Bitget‑hack‑linked trading surge
- ECB Minutes: Higher Long-Term Yields Could Do Part of Its Tightening Work
- Ex-PlayStation Executive Challenges Sony’s Plan to Kill Discs: ‘What Are You Actually Buying?’