The world of cryptocurrency security is entering a new era where the biggest threat is no longer a single massive hack, but the relentless, automated pace of AI-driven financial activity. According to Simone Maini, CEO of blockchain analytics firm Elliptic, the sheer speed and scale at which AI agents can execute transactions and launch attacks could overwhelm traditional compliance teams built for human-paced markets. Speaking at a recent industry event, Maini described this shift as an "automated arms race," where both attackers and defenders are increasingly relying on artificial intelligence to gain the upper hand.
The Rise of AI Agents in Crypto
Artificial intelligence agents—autonomous programs that can analyze data, make decisions, and execute actions without human intervention—are becoming a double-edged sword in the crypto ecosystem. On one hand, they enable legitimate businesses to automate trading, compliance checks, and customer service. On the other, they give bad actors a cheap and efficient way to run large-scale hacks, scams, and fraud operations. Maini noted that the cost of deploying AI tools has dropped significantly, making sophisticated attacks accessible to a wider range of malicious actors.
"The volume of transaction alerts we see is growing exponentially," Maini explained. "Human analysts simply cannot keep up with the pace at which AI agents can move funds across blockchains." This is especially problematic for financial institutions and crypto exchanges that are required by regulators to monitor for suspicious activity. As AI-powered attacks become more frequent, traditional compliance teams face a growing backlog of false positives and missed red flags.
Elliptic’s Agentic Compliance System
In response to this challenge, Elliptic has developed its own AI-driven compliance system. The company recently raised $120 million in a funding round from investors including Nasdaq and Deutsche Bank, signaling strong institutional interest in automated security solutions. The new system uses AI agents to analyze blockchain data in real time, flagging suspicious transactions and identifying patterns that would be invisible to human analysts. Unlike static rule-based systems, these agents continuously learn and adapt to new threats, enabling faster and more accurate detection.
"We are moving from reactive to proactive security," said Maini. "Instead of waiting for a hack to happen and then investigating, our agents can predict and prevent attacks before they cause significant damage." This proactive approach is critical as the number of automated transactions on blockchains like Ethereum and Solana continues to rise. Elliptic's agents are designed to handle the scale of hundreds of millions of transactions per day, far beyond what any human team could monitor.
The Scale of the Threat
To illustrate the magnitude of the problem, consider the growth of decentralized finance (DeFi) protocols. In 2024, DeFi platforms processed over $2 trillion in transactions, with a significant portion involving automated market makers and bot-driven trades. Hackers have also embraced automation: in 2025, flash loan attacks, which often use AI to identify and exploit vulnerabilities, caused losses of over $300 million. These attacks happen in seconds, leaving little time for human intervention.
Maini warned that the next frontier will be AI agents that can simulate entire attack strategies, testing millions of combinations of smart contract parameters before striking. "We are already seeing evidence of AI being used to write malicious code that is harder to detect with traditional signature-based systems," she said. As a result, security firms are investing heavily in machine learning models that can spot anomalous behaviour without relying on pre-defined rules.
Institutional Demand for Automated Compliance
The $120 million raised by Elliptic reflects the urgent need among institutional investors for robust, automated compliance tools. Banks, hedge funds, and asset managers are increasing their exposure to digital assets, either directly or through products like Bitcoin ETFs. However, they remain wary of the regulatory risks associated with money laundering, sanctions evasion, and market manipulation. Automated agentic compliance systems offer a way to manage these risks without requiring massive human teams.
"Institutions want to move fast in crypto, but they need to do it safely," said Maini. "Our system gives them the confidence that they can scale their operations without running afoul of regulators." This is particularly important in the wake of high-profile enforcement actions by the U.S. Securities and Exchange Commission and other global regulators. The ability to demonstrate robust surveillance and reporting capabilities is becoming a competitive differentiator for crypto service providers.
Broader Implications of AI in Security
The trend extends beyond crypto. The convergence of AI and cybersecurity is reshaping the entire digital security landscape. In traditional finance, AI is being used to detect fraudulent transactions and automate threat hunting. However, the decentralized and pseudonymous nature of crypto makes it particularly vulnerable to AI-driven exploitation because there are fewer gatekeepers and lower barriers to entry.
Additionally, researchers and builders are increasingly concerned that AI is accelerating the timeline for quantum computing, which could break current encryption standards. As reported in a related article, the combination of AI and quantum computing poses a threat not just to blockchains but to the internet as a whole. Elliptic's Maini acknowledged this risk but emphasized that the immediate challenge remains the exponential growth of AI agents operating on existing infrastructure. "Quantum is important, but we have a more pressing battle right now with AI-driven crime," she said.
The Road Ahead
As crypto security evolves into an automated arms race, the winners will be those who can deploy the most adaptive and intelligent AI agents. Maini predicted that within the next two years, the majority of compliance alerts will be triaged by AI without human intervention. This will allow human analysts to focus on the most complex and high-risk cases. However, she cautioned that attackers will also improve their AI tools, leading to a continuous cycle of innovation and counter-innovation.
"This is not a problem that will be solved once and for all," she said. "It is an ongoing race that requires constant investment in research and development." For the crypto industry, the message is clear: complacency is not an option. The systems that worked for a market of millions may fail in a market of billions of automated transactions. Only those who embrace the AI arms race will survive and thrive in the coming years.
Source: Coindesk News