AI goes hunting: How agents track crypto criminals

Traders Union
All News
Editors' Picks
How AI helps detect suspicious fund flows faster

From pseudonymity to transparency

From the start, blockchain created an illusion of anonymity. Addresses without names, transactions without banks, the flow of funds — like a chaotic stream of numbers. For a long time, it was genuinely difficult to understand who stood behind a wallet and where the money was going.This pseudonymity made cryptocurrencies a convenient tool for shadow schemes — money laundering, financing illegal services, and bypassing sanctions. The problem was not that the data was unavailable, but that there was too much of it, scattered and unstructured.

The situation began to change with the emergence of clearer rules. The US, Europe, and Asian countries tightened requirements for exchanges, introduced KYC, and implemented monitoring of suspicious transactions. At the same time, blockchain analytics tools evolved, learning to cluster addresses, track fund flows, and link them to real-world services.

As a result, a system long considered practically anonymous is turning into one of the most transparent financial infrastructures. Blockchain has always been a public ledger. Now these traces can also be quickly read, correlated, and attributed.

How the “blockchain reading” market emerged

Amid tightening regulation and the deeper integration of cryptocurrencies into traditional finance, a separate market of tools capable of systematically analyzing blockchain data gradually took shape. If earlier such investigations were the domain of enthusiasts and niche experts, over time they turned into commercial products.

Companies like Elliptic, Chainalysis, and later TRM Labs began building platforms that collect data from multiple blockchains, cluster addresses, track fund flows, and flag risky activity. Importantly, these solutions were never aimed at retail users, but at large clients — government agencies, law enforcement, banks, and crypto exchanges.

Exchanges use such systems to screen transactions and clients, banks to avoid dealing with “dirty” funds, and government agencies for investigations and sanctions enforcement.

Loading...

The tweet was deleted by the author.

But we saved everything 🙂.

Notably, their effectiveness was quickly proven in practice. These tools have been used in investigations of large-scale money laundering schemes, shutdowns of illegal services, and tracking transactions linked to sanctioned jurisdictions.

How AI entered blockchain analytics

AI became the tool that, long before the current boom, helped accelerate blockchain analysis. Back in 2019, Elliptic together with the MIT-IBM Watson AI Lab published a large labeled dataset of Bitcoin transactions to train models aimed at detecting illicit activity.

These approaches addressed the market’s core problem — data volume. When dealing with billions of transactions and complex routes across multiple networks, a human simply cannot process all the information quickly. That is why analysts increasingly relied on models capable of identifying patterns in massive datasets and uncovering connections invisible to manual analysis. In 2024, Elliptic reported a new study based on nearly 200 million Bitcoin transactions, where a model was trained to detect not only individual suspicious wallets but entire money laundering schemes.

Over time, the role of these systems expanded. They began not only to flag risks, but also to help structure investigations: tracking fund flows, suggesting possible links between addresses, and reducing analysis time. Elliptic explicitly stated that such models helped uncover new laundering schemes and previously unknown illicit wallets, with the results already being used to improve their products.

Loading...

The tweet was deleted by the author.

But we saved everything 🙂.

In effect, AI became an “invisible layer” within analytics platforms. But even with these capabilities, a key limitation remained: the system assisted analysts but could not fully replace their work.

When AI stops being just a tool

The market entered a new phase after Chainalysis introduced blockchain intelligence agents in late March 2026. Previously, AI operated inside analytics platforms and mainly helped specialists find links between addresses faster. Now, part of the analytical work itself is being delegated to it. This is no longer just a feature, but an attempt to turn the system into a полноценный investigation assistant.

In practice, this changes how blockchain data is handled. A user formulates a query in natural language, and the system independently selects relevant data, builds the analytical logic, and produces an answer. Chainalysis emphasizes that such solutions rely on billions of transactions and millions of prior investigations — effectively working on an accumulated knowledge base of typical fund flows, risks, and schemes.

Loading...

The tweet was deleted by the author.

But we saved everything 🙂.

The key shift is that the analyst’s role is beginning to change. Previously, a human conducted the investigation end-to-end, while the system only accelerated отдельных этапов. Now, the machine can trace fund routes, structure facts, and compile them into a report for further verification. According to the company, in some cases this already reduces complex investigations from days to minutes.

At the same time, the barrier to entry is changing, as access to analytics gradually expands — not only for narrow specialists and large players, but also for a broader range of market participants who can formulate queries and receive ready-made insights.

In effect, the market is moving from tools that simply accelerate analysis to systems that take over part of the thinking in the process.

When transparency becomes a market rule

For the crypto market, this is not just another wave of technology, but a shift in the rules themselves. If the analysis of addresses and fund flows approaches near real time, freezes become routine rather than exceptional. Liquidity stops being neutral: funds with a clean history move faster, while the rest get stuck in checks before even reaching major exchanges or fiat.

This means that risk data turns into a competitive advantage. Those who detect problematic routes earlier are less likely to lose time to delays, face blocks, or deal with failed settlements. The analysis of fund origin is gradually becoming as integral to trading infrastructure as fees or execution speed.

For the legitimate market, this is largely good news: more predictability, fewer toxic funds, and higher trust from traditional finance. For those operating in gray zones — the opposite. But the key point is different: a market that built its reputation on opacity is becoming increasingly indistinguishable from traditional financial infrastructure. And that may be the most important consequence — not for regulators, but for the crypto market itself.

This material may contain third-party opinions, none of the data and information on this webpage constitutes investment advice according to our Disclaimer. While we adhere to strict Editorial Integrity, this post may contain references to products from our partners.