First AML

Before building the platform, First AML processed more than two million cases as an operator, so its workflow, case management and screening product functioned before any AI arrived. The models add a document reader that parses trust deeds and builds ownership structures, biometric identity matching and a screening agent that triages results.

Compliance rules and review steps belong to the customer's team, and the screening agent labels each result a false positive, false match or true match for a person to accept or override. What counts as routine, or what the agent may close without review, is not stated.

Published figures measure speed, such as a complex document read in around sixty seconds, not correctness. Extraction accuracy and precision or recall for the screening agent's labels are unpublished, though an accepted false positive label makes a real match disappear.

Rampersand, Blackbird and Alceon are among six financial firms named on the Australian pricing page, alongside scale stated as thousands of companies and millions of clients verified. No outcome attaches to any of them, and a 50 to 80 percent cut in due diligence time in its own commentary is unattributed.

Its own commentary floats reinforcement learning on case outcomes and federated learning across institutions as opportunities, yet whether customer data or analyst decisions train shared models is not stated. AI data use is left out of the security statement entirely.

Platform activity logs are kept up to a year and customer specific audit logs for the customer's lifetime, but no subprocessor list, case document retention schedule or processing terms are published. Identity documents, trust deeds and beneficial ownership details of its customers' customers sit in those case records.

ISO 27001:2022 certification with annual audits is stated, and a Drata trust center lists SOC 2, ISO 27001 and SSAE 18 reports beside a public status page. Neither the SOC 2 report type nor report access without a request is shown.

Statutory obligations sit with its customers as reporting entities, and First AML states no license, registration or sandbox of its own. Detailed commentary on New Zealand and Australian regulatory change is subject matter knowledge, not standing.

Commentary names bias as a central risk of AI in this field, yet nothing addresses its own models: no fairness disclosure or evaluation across jurisdictions or entity types. Entity research aims at foreign entities in thin data jurisdictions, where public records are sparse, so reports may be incomplete exactly where buyers lean on them.

A person wrongly listed as a beneficial owner from a trust deed, or wrongly left off, gets no notice that an automated reading produced the entry and no route to correct it. First AML publishes nothing on who bears the cost of an extraction error.

Document understanding, facial biometrics and screening triage name no model provider, family or version. Where inference runs, and whether documents reach an external model provider, goes unexplained.

Aderant's legal practice management software now embeds the compliance workflow through a February 2026 partnership, reaching law firms where they already work. Beyond that, the integration catalog is not enumerated, so which customer or fund administration systems connect is unclear.

Two AWS instances are named by region: Asia Pacific data in Sydney with recovery in Melbourne, and European data in Ireland with recovery in Frankfurt. On premises and single tenant options are not offered, and no formal transfer position is published.

Three tiers are described by the organization each suits, with every price marked on application and varying by jurisdiction. Its entry tier's qualifying volume of 400 individuals or 100 complex entities a year tells a firm whether it fits, but nothing about cost.

New Zealand, Australia, the United Kingdom and Europe each have a site, across accounting, law, real estate and finance. Finance is listed last of four verticals, and the base leans toward professional services firms with anti money laundering duties.