Sources & Methodology

The rules we cite throughout come from two Interpol instruments, both published on interpol.int. The Constitution supplies Article 3 — the absolute prohibition on the Organization undertaking any intervention or activities of a political, military, religious or racial character — which the CCF applies through the “predominance test,” weighing the factors set out in Article 34(3) of the Rules on the Processing of Data. The Rules on the Processing of Data (RPD) supply the rest: Article 83’s conditions for Red Notices, including the exclusion of offences arising from private or administrative disputes under Article 83(1)(a)(i) and the penalty thresholds (an offence carrying a maximum of at least two years for prosecution cases, or at least six months imposed or remaining for conviction cases); Article 12’s data-quality requirements that information be accurate, relevant, not excessive, and up to date; and the procedural articles governing the CCF itself, including the Article 40(1) four-month target for deciding requests.

Every CCF caseload and outcome figure on this site comes from the Commission’s own annual activity reports, chiefly the report covering 2024, read against prior years for trend. The specific figures we use: 2,586 new admissible requests in 2024 — a record, roughly 45% deletion requests, 37% access, 18% revision — against 1,265 in 2018, a caseload that has more than doubled. 272 of 539 deletion requests decided on the merits in 2024 were found non-compliant — the source of the “roughly half” framing we use, with the compliance-finding trend rising from about 26% (2022) to 32% (2023) to 40% (2024). A further 164 requests in 2024 were resolved by deletion before a Commission decision, typically because the source country withdrew or the General Secretariat acted. On timing, the same reporting shows about 30% of deletion requests exceeding nine months in 2024, up from 15% in 2023 — the honest counterweight to the four-month target.

Figures about the notice system itself come from Interpol’s published data and annual reporting: approximately 86,000 Red Notices in circulation, of which fewer than 10% are public extracts on the Interpol website — the basis for our repeated caution that a clear public search proves little. 15,548 Red Notices were published in 2024, a record and roughly 27% above the prior year. Interpol’s internal screening body, the Notices and Diffusions Task Force (NDTF), refused or cancelled 2,462 notices and diffusions in 2024, up 54% from 1,598 in 2023 — evidence both that screening catches real abuse and that abuse arrives at scale. The Organization counts 196 member countries, and it has suspended a member’s access only once in its history (Syria) — a fact we cite when discussing the transparency critique, since no public country-by-country refusal data exists.

Our pages on the refugee ground rest on the 1951 Refugee Convention’s non-refoulement principle and on Interpol’s own policy, in place since 2015, under which a Red Notice or diffusion sought by a person’s country of origin against a recognised refugee is, as a general rule, treated as non-compliant with Interpol’s rules. We also note the documented limits: the protection turns on recognised status and unchanged risk, and cases discuss its persistence — or complication — after naturalisation. Where we describe how the CCF weighs asylum grants as evidence of political character more broadly, that reading comes from the Commission’s published decision extracts and its reports’ discussion of Article 3 practice.

The country pages attribute their central claims to named external sources rather than asserting them as ours. The identification of China, Russia, and Turkey as the most prolific abusers draws on evidence submitted to the UK Parliament’s Joint Committee on Human Rights inquiry into Interpol misuse (2026), with practitioners’ evidence often placing Turkey and Russia at the very top. The January 2026 whistleblower leak, reported by the BBC with the investigative outlet Disclose and addressed in an Amnesty International statement, revealed the breadth of politically tainted notices inside the system and renewed calls for transparency and suspensions. Country specifics — Russia’s share of public notices and its “corrective measures”; Turkey’s post-2016 mass filings and SLTD passport cancellations; China’s Fox Hunt and Sky Net operations and documented family coercion; the UAE’s $54 million 2017 donation to Interpol, roughly equal to all other members’ contributions combined — come from the inquiry evidence, court records, NGO documentation, and established investigative reporting cited on those pages.

Descriptions of how filing actually works — that the CCF accepts requests directly from individuals at no charge; that its admissibility stage is an identity-and-clarity check while the Requests Chamber examines the merits; that its decisions are binding on Interpol, with deletion executed by the General Secretariat and a certificate of deletion available; and the current portal’s mechanics, including one request type per submission, the separation of access requests from deletion requests, and the limit of twenty appendices — come from the CCF’s published Statute, operating rules, and application guidance on the Interpol website. Where the portal’s behaviour differs in practice from the written guidance, our pages describe the practice and say that is what we are doing.

Three habits govern every page. Verify: a claim enters the site only when it traces to a primary document or, failing that, to a named institution or established outlet — and the pages tell you which. Date: every content page carries a “last reviewed” date, and statistics are tied to their year in the text, because this system moves annually and an undated number is a future error. Attribute: contested characterisations — that a state abuses the system, that a case was political — are attributed to the inquiry, court, or organisation that made them. We write plainly about patterns, but the reader can always see whose finding is being reported.

Three source types are excluded by policy. Law-firm marketing content — the bulk of what ranks in this field — is never treated as authority, because its incentives run against the reader’s. Anonymous or single-sourced claims about individual cases are not repeated, however vivid, unless corroborated by records or established reporting. And we do not present Interpol’s or any government’s self-description as the last word on its own conduct: official sources are authoritative for rules and their own statistics, while their performance is assessed against external evidence — the same standard we apply to ourselves by publishing this page.

When a figure changes — a new CCF annual report, an amended rule, a policy shift — the affected pages are updated and their review dates advanced. Material errors are corrected as such, not silently patched: if we had a number wrong, the fix is a correction, and we treat reader reports as the most valuable input we get. To report an error, use the contact page with the page URL and, ideally, the source that contradicts us. A resource in a field where mistakes cost people liberty has exactly one sustainable policy on being wrong: quickly, openly, and gratefully corrected.