AI Reshapes the Crypto Crime Landscape

Recent monitoring data from TRM Labs reveals a sharp acceleration in the criminal adoption of artificial intelligence. Over the past year, the use of AI in cryptocurrency-related illicit activities has jumped by 40%. The firm's ‘AI Crime Adoption Index’ now sits at 54 points, placing it firmly in the ‘Emerging’ phase—a significant leap from its score of around 28 in early 2024.

Scams: The ‘Mature’ Frontier of AI Crime

Among all crime categories leveraging AI, scams stand alone as having reached a ‘Mature’ stage of adoption. A critical trend underpins this: since 2022, the proportion of reported crypto scams involving technologies like deepfake videos, AI voice cloning, and chatbots has grown approximately 13-fold.

The data for 2026 is even more alarming. Losses from reported deepfake scams in the first part of this year alone are already 263% higher than the total recorded for all of 2025. This explosive growth signals that bad actors have moved beyond experimentation to operational efficiency.

State Actors Weaponize AI Tools

The report highlights a troubling escalation: nation-state hackers are systematically integrating AI into their operations. Specifically, groups linked to North Korea have been documented:

  • IT Workforce Infiltration: Using deepfakes to impersonate recruiters or tech executives, gaining trust to plant malware or steal credentials.
  • Enhanced Social Engineering: Leveraging AI to analyze public data and craft highly personalized phishing messages, dramatically increasing success rates.
  • Automated Vulnerability Discovery: Deploying AI tools to continuously scan for weaknesses in networks and smart contracts, speeding up attack preparation.

This state-sponsored, systematic weaponization of AI marks a new phase in cyber threats, where attackers benefit from vastly improved efficiency and precision.

A Fundamental Shift in Defense is Required

Combating AI-powered crime demands moving beyond traditional blocklists and pattern matching. Experts are calling for more adaptive security frameworks, including:

  • Implementing multi-factor authentication processes robust against deepfake impersonation.
  • Training staff to recognize social engineering attacks in the AI era.
  • Developing specialized tools capable of detecting AI-generated content and behavioral anomalies.

The dual-use nature of technology is on full display. As security teams use AI to predict and prevent attacks, adversaries are using the same technology to forge sharper spears. This arms race is only accelerating.