When US Customs and Border Protection stops a container under the Uyghur Forced Labor Prevention Act (UFLPA), it does not ask the importer for a supplier code of conduct. It asks for evidence tied to that specific consignment — which factory, which batch, which date of production — and the burden of proof sits with the importer, not the agency. That single shift, from "describe your supply chain" to "prove this shipment", is the reason forced-labour compliance has quietly become a data problem as much as a legal one.
Customs wants a consignment record, not a policy document
Under the UFLPA, goods that CBP believes were mined, produced, or manufactured wholly or in part in China's Xinjiang Uyghur Autonomous Region, or by an entity on the UFLPA Entity List, are subject to a rebuttable presumption: they are presumed to have been made with forced labour and are barred from entering the United States unless the importer rebuts that presumption. To do so, CBP's own guidance requires "clear and convincing evidence" that the specific goods in question were not produced wholly or in part by forced labour — a high evidentiary bar, and one that has to be met at the level of the detained shipment, not the level of a general supplier relationship. A signed code-of-conduct letter from a factory says nothing about the batch sitting in a bonded warehouse; a record showing which factory produced that batch, and when, at least gives the importer something concrete to submit.
The EU is building the same expectation into law on a slower clock. Regulation (EU) 2024/3015, the EU Forced Labour Regulation, entered into force on 13 December 2024 and applies in full from 14 December 2027, with a handful of provisions — including the designation of competent authorities and the risk database — coming into effect earlier. Once it applies, it prohibits placing on, or exporting from, the EU market any product made with forced labour at any stage of its production, and gives national authorities and the Commission the power to investigate and order the withdrawal of specific consignments, not just to fine a company in the abstract. The Commission published its implementing guidelines on 30 June 2026 and launched a public risk database and single information submission point on 26 June 2026, both intended to help authorities and businesses identify where the risk actually sits — sector, geography, product type — rather than treating "supply chain risk" as a single undifferentiated line item.
Why a supplier attestation doesn't survive a rebuttable presumption
Most forced-labour compliance programmes were built to answer a different question than the one now being asked. A supplier attestation, an audit report, or a code-of-conduct sign-off describes a relationship at a point in time — it says "this factory told us it doesn't use forced labour" or "an auditor visited on this date." None of that identifies which specific units in a specific shipment came from which specific production run. When the presumption of guilt sits with the importer, as it does under the UFLPA, the importer needs evidence that maps to the goods actually being examined, not evidence that maps to the supplier relationship in general. A brand that can only describe its supply chain at supplier level has nothing to hand over when customs asks about a pallet; a brand with a batch and factory reference tied to each unit at least has something to submit for review.
The commercial cost lands long before any court does
A detained shipment is a working-capital problem before it is a legal one. Stock sits in bonded storage, unsellable, while the brand pays to store it and loses the selling season it was bought for. Retail partners who committed shelf space or a launch date don't wait for a legal resolution — a missed delivery window is a commercial failure regardless of why the goods didn't arrive, and repeated disruption is grounds for a retailer to delist a supplier rather than risk its own exposure. None of this requires a finding of forced labour to hurt; the disruption itself is the cost, and it falls on the brand long before any penalty, if one ever comes, is decided.
What item-level provenance can actually show — and the limit that matters
A digital identity tied to the individual item or batch — the kind SmartLinks builds via a QR code or NFC tag resolving to a record the brand controls — can carry the specific facts a rebuttal or an investigation asks for: which factory produced this unit, which batch it belongs to, and the date of manufacture, alongside whatever certificates or supplier documents the brand has chosen to attach. That is a materially better starting point than a supplier-level statement, because it answers the question actually being asked: not "what is your policy", but "what do you know about this consignment."
It is also, deliberately, not more than that. SmartLinks stores and displays the provenance data a brand puts into it. It does not audit the factory, inspect labour conditions, or verify that what a supplier declared is true. A passport makes a brand's claims specific and traceable to a batch and a date; it does not itself validate those claims, any more than a filing cabinet validates the documents inside it. A brand that describes its own passport as proof of clean labour practices is overclaiming, and that overclaim is more reputationally dangerous than having no system at all — a forced-labour accusation that catches a brand claiming something its own record can't support is a worse story than one where the brand simply didn't have the data yet.
Two different bars, and neither is what most statements currently meet
The UK's Modern Slavery Act 2015 sets a lower bar than either the EU or US regimes. Section 54 requires commercial organisations with a total turnover of £36 million or more to publish an annual slavery and human trafficking statement; it does not require them to carry out due diligence, and the Act explicitly allows a statement to say the organisation has taken no steps at all, provided it says so. The only sanction for non-compliance is an injunction, which in practice has gone unused. That is a transparency duty, not a due diligence duty, and it is a different and much lighter obligation than either the UFLPA's evidentiary presumption or the EU Forced Labour Regulation's market ban.
The EU's Corporate Sustainability Due Diligence Directive (CSDDD) does impose a genuine due diligence duty, but its scope has narrowed since it was first adopted. Following the "Omnibus I" amending directive, the CSDDD now applies to EU companies with more than 5,000 employees and over €1.5 billion in global net turnover — up from the original 1,000 employees and €450 million — with member states required to transpose the rules by 26 July 2028 and the directive applying from 26 July 2029. That pushes CSDDD-specific due diligence obligations several years out and narrows them to the very largest companies; it should not be conflated with the Forced Labour Regulation's market ban, which applies to any product regardless of company size, or with the UFLPA, which is a US import control with no EU equivalent yet in force.
Start with the record you'll actually be asked for
None of these regimes are asking brands to prove a negative. They are asking for a specific, retrievable answer about a specific consignment: which factory, which batch, when. Most brands currently keep that information, if they keep it at all, in a supplier file that nobody can search by shipment. Building an item- or batch-level digital record now — starting with the products most exposed to Xinjiang cotton, minerals, or high-risk sourcing regions — means that when a shipment is stopped, the answer already exists rather than needing to be assembled under a customs deadline. SmartLinks is free until you go live, so the sensible place to start is the highest-risk product line, not a company-wide rollout planned for next year.
See how SmartLinks handles ESPR & DPP →
Explore SmartLinks Hub →
Book a walkthrough with our team →
Sources: Regulation (EU) 2024/3015 (EU Forced Labour Regulation) — https://eur-lex.europa.eu/eli/reg/2024/3015/oj
Modern Slavery Act 2015, section 54 — https://www.legislation.gov.uk/ukpga/2015/30/section/54
US Customs and Border Protection, Uyghur Forced Labor Prevention Act guidance — https://www.cbp.gov/trade/forced-labor/UFLPA