Ian Rogers, chief human agency officer at Ledger, said artificial intelligence tools are making cryptocurrency hacks faster and easier than ever before [1, 2].
The shift represents a critical security risk for digital asset holders. As technical barriers drop, the volume and speed of attacks may increase, potentially outpacing traditional security responses.
Speaking on Bloomberg’s “Crypto” program in July 2026, Rogers said AI is lowering the barrier to entry for attackers [1]. He said hackers can now use advanced language models to generate exploit code in minutes [2].
Rogers said the technology is being used to automate wallet attacks, which makes the process both faster and cheaper [1]. This automation allows attackers to target vulnerabilities with a level of efficiency previously reserved for highly skilled programmers.
Industry data supports these concerns. Some security researchers report that AI can scan thousands of smart contracts per second to find vulnerabilities [3]. This capability allows attackers to identify weak points across a vast ecosystem in a fraction of the time it would take a human auditor.
These warnings follow a period of heightened volatility in the sector. A wave of high-profile cryptocurrency hacks occurred in April 2026 [4]. One security researcher said the current state of AI-driven attacks is turning the hacking process into a "point-and-click operation" [5].
Despite these warnings, some companies are integrating AI into their own security frameworks. Ledger has announced an AI toolkit designed to allow AI agents to read account balances [6]. However, the company said these agents cannot move funds without a human-signed hardware approval [6]. This creates a tension between the perceived threat of AI-driven exploits and the industry's push to adopt AI for utility.
“AI is lowering the barrier to entry for attackers.”
The democratization of exploit generation via AI shifts the crypto security landscape from a battle of skill to a battle of automation. While hardware-based approvals remain a primary defense, the ability of AI to scan thousands of contracts per second suggests that software-level vulnerabilities will be discovered and exploited almost instantaneously, necessitating a move toward AI-driven real-time defense systems.
