Charles Guillemet, Chief Technology Officer at cryptocurrency wallet provider Ledger, has issued a stark warning: artificial intelligence is dismantling the security defenses of the crypto industry. He noted that the proliferation of AI tools has drastically reduced the cost and complexity of launching cyberattacks, pushing them toward near-zero levels.
“Finding and exploiting vulnerabilities has become incredibly easy,” Guillemet said in an interview. He argues that AI is fundamentally shifting the economics of cybersecurity, allowing attackers to reap massive rewards for minimal investment.
AI Escalates Crypto Hacking Threats
Recent security incidents in the crypto space underscore this trend. This week, the Solana-based decentralized finance protocol Drift was hit by an attack, resulting in losses of $285 million. Just a week prior, the yield protocol Resolv lost $25 million to a similar exploit. According to DefiLlama, losses from hacks and exploits in the crypto sector have exceeded $1.4 billion over the past year.
In traditional cybersecurity, the defensive strategy relies on “asymmetry”—the idea that the cost of attacking a system should be significantly higher than the potential gain. However, Guillemet believes AI is eroding this advantage. Software reverse engineering or the creation of exploit chains, which once took professional security researchers months to complete, can now be achieved in seconds using AI prompts.
Furthermore, AI-generated code introduces new risks. As developers increasingly rely on AI-assisted programming tools, hidden security vulnerabilities can be injected into code on a massive scale. Guillemet expressed concern over this shift: “We are churning out vast amounts of inherently insecure underlying code because there is currently no ‘one-click security’ button.”
To combat this emerging threat, Guillemet urges crypto protocol developers to fundamentally re-architect their security frameworks. He specifically emphasized the importance of “formal verification”—using mathematical proofs to verify the correctness of code logic rather than relying solely on traditional audits. He also recommends that the industry shift toward hardware-based security solutions and offline storage, while reminding users that they should be mentally prepared for the possibility that systems may ultimately fail.
