Blockchain analytics firm Chainalysis announced on Tuesday a significant advancement in its platform with the introduction of 'blockchain intelligence agents.' These are specialized, autonomous AI tools designed to assist law enforcement agencies, compliance teams, financial institutions, and crypto businesses in investigating, monitoring, and combating illicit activity on blockchains more efficiently.
The agents are trained on Chainalysis' proprietary dataset, which the company has built over more than a decade in business. This dataset includes insights from billions of screened transactions and over ten million investigations. "This isn't a new product or a bolted-on chatbot feature. Agents are the evolution of the platform we've built and everything we've learned," said Chainalysis CEO Jonathan Levin in a statement.
According to the announcement, the AI agents can operate in two modes: a deterministic workflow where the same inputs always produce the same outcome, or an exploratory mode. Both settings generate detailed audit trails, and human operators can set the agents' level of independence and specific tasks to automate. These audit trails provide a step-by-step record of the data used, the agent's reasoning, and the actions taken, which is crucial for meeting compliance and legal standards.
Chainalysis emphasized its unique position, noting it is the only onchain analysis firm whose data has been ruled reliable and admissible in court. Levin added, "We have the context of every type of investigation you run and the compliance obligations you navigate daily. We know how transactions on the blockchain actually function, and the workflows, audit trails, and standards of evidence you rely on."
The firm plans to begin rolling out these tools in the summer of 2026. During the testing phase, agents have already been deployed for tasks including open-source intelligence collection, tracking complex investigations across multiple blockchains, and providing raw alerts. The bots also possess capabilities similar to general-purpose AI, such as coding web applications and generating summary reports.
Levin framed the development as a necessary countermeasure, stating, "As bad actors increasingly leverage AI to scale their operations, it's critical that those working to stop them do the same."
