Sweden rewrites its periodic AML reporting for 2027

Finansinspektionen has announced a full overhaul of the periodic anti-money laundering report. The current question set is retired and replaced with questions built on the risk indicators developed jointly by the EBA and the new EU Anti-Money Laundering Authority, AMLA — the EU’s common methodology for risk-classifying supervised firms.

The change takes effect on 1 January 2027, with data as of the balance date 31 December 2026. The reporting window stays 1 January – 31 March, and submission remains through FI’s Fidac portal. Every firm under FI’s supervision that falls within the Anti-Money Laundering Act is in scope.

What sits underneath

The new questions are not FI’s invention. They come from AMLA’s draft technical standard under Article 40(2) of the new AML directive, which defines a single EU methodology for assessing every financial-sector firm: an inherent risk score across four categories — customers, products and services, geographies, distribution channels — a controls quality score, and a residual risk rating on a common four-step scale. To feed it, the standard specifies roughly 100–150 datapoints per entity in its annex: transaction volumes by country, customer counts per risk category, crypto exposure, and more. Assessments run annually, with a lighter three-year cycle for the smallest, lowest-risk firms, and the standard applies from 31 December 2027.

In other words: periodic AML reporting is becoming a defined datapoint model with an EU-wide methodology on top — the same construction the EBA has run for prudential reporting for a decade.

The direction of travel

AMLA is already testing the machinery. In March it launched a data collection exercise with selected institutions across the EU — an Excel template carrying the annex datapoints, submitted by late April — to calibrate the risk models that will pick up to 40 entities for AMLA’s direct supervision from 2028. Those directly supervised firms will not answer FI’s questionnaire at all; per FI, they will report according to an EBA taxonomy, the same structured, machine-validated format that already carries COREP OF and FINREP.

The trajectory is hard to miss: Excel template today, EBA taxonomy tomorrow, and a national questionnaire rewritten to match the same indicators in between. The EU’s supervisory architecture keeps converging on one model — a data point model, a taxonomy, validation rules.

That is the model Sigill is built on. When a reporting obligation moves from free-form questions to defined datapoints under an EBA taxonomy, it lands on ground we already operate: render the template from the model, validate against the official rules, file a package the supervisor’s systems accept on the first attempt.

It is also why we treat national returns as part of the job rather than an afterthought. The Swedish AML report is a national obligation today and an EU-methodology obligation tomorrow, and an institution’s reporting is only complete when both are covered from the same set of figures.

FI expects to publish implementation details in autumn 2026. We are following them, and Sigill’s framework support will track the new requirements as the technical standards land.

Sources: FI: Omfattande ändringar i periodisk rapportering penningtvätt · AMLA: Final report, draft RTS under Art. 40(2) AMLD · AMLA: Data collection exercise