- AI is increasingly being used by hackers to exploit vulnerabilities in DeFi, altering the security landscape.
- April was particularly devastating for DeFi protocols, with over $690 million lost to attacks.
- Smart contract audits are improving, prompting hackers to shift focus to operational security and infrastructure.
- The role of AI in cyberattacks is debated, with some studies downplaying its impact on creating “super hackers.”
DeFi Security Under Siege: The Role of AI in Recent Attacks
The decentralized finance (DeFi) sector is grappling with a new wave of security challenges as hackers increasingly leverage artificial intelligence (AI) to find and exploit vulnerabilities. According to CertiK’s co-founder and CEO, Ronghui Gu, this shift has resulted in a significant increase in attacks, with April witnessing only three days free from hacking incidents.
Unprecedented Losses Due to DeFi Exploits
In April 2026 alone, DeFi protocols suffered losses exceeding $690 million due to various attacks. This month marked the worst for DeFi in terms of financial losses since March 2022, excluding a major breach at Bybit exchange in February 2025. Analysts reported 28 separate incidents resulting in over $635 million lost. Notably, breaches involving Drift Protocol and KelpDAO accounted for almost $600 million of these losses.
A Shift Towards Infrastructure Attacks
With smart contract audits becoming more robust, hackers are shifting their focus towards exploiting weaknesses in operational security and infrastructure. Gu emphasized that most crypto projects still underestimate these risks. For instance, the attack on Drift Protocol was linked to administrative control seizures allegedly orchestrated by North Korean hackers.
Rising Concerns Over AI-Driven Cyber Threats
Despite CertiK’s warnings about enhanced hacker capabilities through AI, some researchers urge caution against overestimating this threat. A study analyzing discussions on darknet forums found no evidence of widespread emergence of AI-based “super hackers.” Instead, malicious actors primarily use neural networks for SEO spam and phishing campaigns.
International Warnings and Industry Responses
International bodies have raised alarms about potential systemic risks posed by advancing AI technologies. The International Monetary Fund highlighted concerns over models like Mythos from Anthropic potentially causing financial disruptions through cyber threats.
In response, leading tech companies are initiating their own cybersecurity projects utilizing AI. OpenAI’s Daybreak project aims at identifying vulnerabilities and integrating cybersecurity measures directly into software development processes.
The Need for Industry Collaboration
Ronghui Gu stresses the importance of industry-wide collaboration to effectively respond to modern crypto breaches. The coordination required between protocols, blockchains, and exchanges makes tackling these challenges increasingly complex.
CertiK also highlights the growing threat from North Korean hackers who have reportedly stolen over $6.75 billion worth of crypto assets since 2016 through numerous documented incidents.
The evolving tactics employed by cybercriminals demand that the cryptocurrency industry remains vigilant and proactive in fortifying its defenses against emerging threats powered by technological advancements like artificial intelligence.