Attribution

Attribution is linking a blockchain address to the real-world entity behind it, such as an exchange, a scam, a mixer or a sanctioned person. AML checkers rely on these labels to work out where money came from. An AML score is only as good as the attribution data behind it.

What attribution means

A blockchain shows addresses and amounts, not names. Attribution is the work of putting a name or a category on an address: “this is a deposit address of exchange X”, “this wallet collected payments for a phishing scam”, “this address belongs to a sanctioned group”. A labelled address is called attributed; the rest are unknown.

Vendors present it as a core part of their product. TRM Labs says its Wallet Screening returns a list of risks and attribution for an address.[1] GetBlock’s documentation describes a report with a score, an exposure breakdown, compliance flags and attribution where available.[2]

Where the labels come from

  • Official lists. OFAC can name crypto addresses as identifiers on SDN entries, and says these lists may not be exhaustive.[3] It first did so in November 2018.[4]
  • Public sources. Court documents, hack reports, scam reports from victims and an entity’s own announcements.
  • Direct observation. Analysts deposit to or withdraw from a service and record the addresses it uses.
  • Clustering. Once one address is known, address clustering can extend the label to other addresses the same entity controls. OFAC’s guidance itself notes that unlisted addresses sharing a wallet with a listed one may pose sanctions risk.[5]

How attribution shows up in an AML check

Attribution is what turns a graph of transfers into a risk score. If the coins you are about to receive came from an address labelled “stolen funds”, the checker can count that as exposure. Chainalysis, for example, says its Address Screening shows exposure to sanctioned entities, scams, darknet markets and other categories.[6] AMLBot groups its sources into “danger”, “suspicious” and “trusted”.[7] Each vendor builds its own label database, so two checkers can disagree about the same address. A brand-new address may simply have no label yet; TRM, for example, marks unprofiled addresses as “Unknown”.[8]

What it means for an ordinary user

A low score is not a guarantee: it reflects what one vendor knows today, and labels are added over time. A high score can also be a false positive if a label is wrong or too broad. When the stakes are high, check the address with two services and look at the categories behind the score, not only the number. Our rankings compare checkers by data source, and the risk score guide explains how to read a report.

General information, not legal or financial advice. Rules differ by country and change; check the linked sources.

FAQ

Why do two AML checkers give the same address different results?

Each vendor has its own attribution database and its own scoring rules. One may have labelled an address that the other has not.

Can attribution be wrong?

Yes. Labels are built from lists, reports and heuristics, and they can be incomplete or mistaken. Treat a single result as a signal, not a verdict.

Does attribution reveal my name?

Not normally. It labels addresses by service or category, such as an exchange or a scam. Names are held by exchanges under KYC, not by AML checkers.

Sources

  1. TRM Labs (vendor) — Wallet Screening. Checked 19 Nov 2026.
  2. GetBlock docs — How risk scoring works. Checked 19 Nov 2026.
  3. U.S. Treasury, Office of Foreign Assets Control (OFAC) — OFAC FAQs: Questions on Virtual Currency (topic 1626) (19 Mar 2018). Checked 19 Nov 2026.
  4. U.S. Department of the Treasury — Treasury Designates Iran-Based Financial Facilitators of Malicious Cyber Activity and for the First Time Identifies Associated Digital Currency Addresses (sm556) (28 Nov 2018). Checked 19 Nov 2026.
  5. OFAC, US Treasury — Sanctions Compliance Guidance for the Virtual Currency Industry (PDF) (Oct 2021). Checked 19 Nov 2026.
  6. Chainalysis (vendor) — Address Screening. Checked 19 Nov 2026.
  7. AMLBot — What do we analyze?. Checked 19 Nov 2026.
  8. TRM Labs — June 2025 product highlights (Jul 2025). Checked 19 Nov 2026.