SyRI (Netherlands)

SyRI, short for System Risk Indication, was a Dutch state system for detecting welfare, tax, and labor law fraud. A foundation, the Inlichtingenbureau, linked records from many government databases and compared them against a secret risk model. Matches became risk notifications that the people flagged generally could not know about.

How it worked

The Ministry of Social Affairs and the participating administrative bodies first defined the risk model. Its indicators were kept secret. One disclosed example was unusually low running water use, taken to suggest that a benefits recipient lived elsewhere.

The Inlichtingenbureau pseudonymized the records, putting a code in place of each name, and linked them. The databases covered employment, income, benefits, taxes, fines, property, housing, education, pensions, debts, permits, and more. In principle the linkage could include special-category data, the most sensitive kinds of personal data under European data protection law.

When a person's records matched the model, the code was replaced by the person's name again, and the match became a yes-or-no risk notification. A register held them for up to two years. Before release, the Ministry of Social Affairs screened flagged notifications for obvious false positives: flags on people who were not committing fraud. A released notification could lead to an administrative or criminal investigation, and possibly a home visit.

What the people flagged could know

The law gave no duty to tell people that their data had been processed, or that a risk notification had been filed. So a flagged person generally could not know about, access, or contest a notification. Meanwhile a notification in the register could be re-linked into other agencies' files.

Where it was used

SyRI gained a legal basis in 2014, in articles 64 and 65 of the SUWI Act, the Dutch social security implementation act. From 2014 to 2019, five municipalities requested analyses, SyRI runs over the records of chosen neighborhoods. Only two projects were carried out.

The case file records SyRI's use only in low-income neighborhoods with many migrant residents: in Eindhoven, Haarlem, Capelle aan den IJssel, and Rotterdam-Zuid. It does not say which of these were the two projects carried out. The municipality halted the Rotterdam-Zuid project, in Bloemhof and Hillesluis, in mid-2019. It did so over an unresolved dispute about the project's legal basis, before the planned home visits took place.

What it found

In June 2019, the newspaper De Volkskrant reported that SyRI had not detected a single fraud case since its introduction. In one Capelle aan den IJssel project, 62 of 113 risk notifications were reported to be false positives. After that analysis effectively yielded nothing, the participating parties re-linked their own files independently.

These figures come from court filings and the press, not from an independent government audit.

Who challenged it

A coalition of civil society organizations sued the Dutch State. Among them were the Nederlands Juristen Comite voor de Mensenrechten (NJCM), Platform Bescherming Burgerrechten, Privacy First, Stichting KDVP, the FNV trade union confederation, and the Landelijke Clientenraad. The authors Tommy Wieringa and Maxim Februari joined them. The Public Interest Litigation Project (PILP) coordinated the case.

Philip Alston, the UN Special Rapporteur on extreme poverty and human rights, intervened in the case. His brief to the court was dated September 26, 2019, ahead of the main hearing on October 29. He called dragnet systems like SyRI "the digital equivalent of fraud inspectors knocking on every door in a certain area."

SyRI received the Big Brother Award on November 29, 2019.

What the court ruled

On February 5, 2020, the District Court of The Hague ruled that the SyRI legislation violated Article 8 of the European Convention on Human Rights. Article 8 protects private life. The court held that the scheme failed the test of proportionality, or fair balance. It lacked transparency: the model and its indicators were secret, and the people affected got no notice. It also breached the principles of purpose limitation and data minimization in the GDPR, the EU's data protection law.

The court found that the State bears a special responsibility when it applies new technologies. It flagged risks of discrimination, stereotyping, and stigmatization. It noted that a risk notification had a significant effect on a person, even without formal legal effect.

The court declared the legislation to have no binding effect and ordered its use stopped. On April 23, 2020, the government announced that it would not appeal. The ruling is widely described as one of the first times a court anywhere halted a digital welfare fraud technology on human rights grounds.

What came after

A broader data sharing law, the Wet gegevensverwerking door samenwerkingsverbanden (WGS), entered into force on March 1, 2025. Critics dubbed it "Super SyRI", and a civil rights coalition opposed it. So the question SyRI posed is live, not historical: on what published basis may a state link everyone's records and score the poor?

The question this case asks

SyRI is the pre-harm case in this collection. The control that stopped it came before its individual harms were counted. It came on transparency and privacy grounds, not from measuring who was wrongly flagged.

A court could see the defect on the face of the law. The case file names that defect: a secret model, no notice to the people it flagged, and linked data far beyond SyRI's stated purpose.

Two nearby cases differ. In Rotterdam, an audit opened up a live model and measured its bias. The Dutch childcare benefits scandal was reckoned with through a parliamentary inquiry, years into the damage.

The case file draws a lesson from this. Transparency is not a reporting nicety that comes after accuracy. It is the precondition for correction to exist at all. A system no one can see is a system no one can fix.

What the available tools can and cannot address

A failure pathway is a link between two parts of the network where a mistake made by one part can be passed on to the other. A pathway is closed when mistakes stop passing along it.

Explore (No Targets) sets no targets. Under Service Targets Only, the targets can be met with one tool: Vet connections, for 3 of this case's 12 budget units. It closes four pathways. Two are the linked records and the past notifications compared with the risk model. The others are the notifications copied to agencies and the one indicator set applied to every record.

Service and Safety Targets and All Governance Targets both ask you to close every failure pathway. The targets can be met under both. The cheapest combination costs 11 of the 12 units: Understand the system, Assign a challenger, Escalate checks, Vet connections, and Store less data. While Understand the system is on, three of the others cost less.

Under All Governance Targets, Vet connections and Store less data work at reduced strength unless Understand the system is on.

More tools are not better here. The service target asks that the network keep doing useful work. Every tool at once, ignoring the budget, closes every pathway but misses the service target at all three target levels. Some of the tools take something from the service. Together they take it below its target.