LuxembourgMedia

How We Build and Maintain LuxembourgMedia.com

Last updated 1 June 2026

LuxembourgMedia.com covers Luxembourg's financial sector through a combination of public regulatory data, official statistics, and editorially reviewed content. This page explains exactly where our information comes from, how it is processed, how often it is updated, and where its limitations lie. We believe you deserve to know how a site that covers regulated entities and financial topics actually works.

Our Primary Data Sources

The CSSF Supervised-Entities Register

The Commission de Surveillance du Secteur Financier (CSSF) publishes a public register of entities it supervises — investment funds, management companies, credit institutions, payment institutions, and other regulated actors operating in or from Luxembourg. This register is the authoritative, primary source for the approximately 3,417 entity profiles you will find on this site.

We access this register through CSSF's publicly available data. Our automated pipeline retrieves the register on a weekly cycle, typically processing updates over the weekend so that any additions, removals, or status changes published by the CSSF are reflected on our site within approximately seven days of appearing in the source data. We do not scrape behind authentication walls or access any non-public CSSF systems. Everything we draw on is openly published by the regulator for public use.

Important: LuxembourgMedia.com is an independent directory and editorial platform. We are not affiliated with the CSSF, we are not endorsed by the CSSF, and we have no special data relationship with the CSSF beyond reading the same public register that any member of the public can access.

The GLEIF LEI Database

The Global Legal Entity Identifier Foundation (GLEIF) maintains the global LEI (Legal Entity Identifier) database, which maps legal entities worldwide to standardised identifiers and includes names, jurisdictions, registration details, and corporate relationship data. GLEIF publishes this data under a Creative Commons CC0 1.0 Universal licence, meaning it is released into the public domain with no restrictions on use.

Where a Luxembourg entity has an LEI, we incorporate GLEIF data to enrich profile pages — for example, to confirm legal names, display entity status, or surface parent-subsidiary relationships. Like CSSF data, GLEIF records are pulled into our pipeline on the weekly automated cycle.

We are not affiliated with or endorsed by GLEIF.

STATEC Statistical Data

STATEC is Luxembourg's national institute of statistics and economic studies. We use STATEC's publicly published aggregate statistics — on topics such as the size of the financial sector, employment figures, and macroeconomic indicators — to provide context in our editorial guides and sector overviews. STATEC data is used at an aggregate, informational level and is not used to make claims about individual entities.

We are not affiliated with or endorsed by STATEC.

The Weekly Automated Refresh Cycle

All three data sources — CSSF, GLEIF, and STATEC — feed into a single automated pipeline that runs on a weekly schedule. The pipeline:

This means our data is not real-time. There is an inherent lag of up to approximately seven days between a change being published in a source register and that change appearing on LuxembourgMedia.com. For time-sensitive regulatory matters — verifying whether a specific entity is currently authorised, for example — you should always consult the CSSF register directly at cssf.lu.

How We Use AI to Generate Content

Writing short, useful introductory descriptions for over three thousand entity profiles, plus guides aimed at expats and cross-border workers, is not a task that human editors can accomplish cost-effectively at that scale while keeping content current. We are transparent about how we handle this.

What is AI-generated

We use the Claude API (developed by Anthropic) to generate:

AI generation is used as a drafting tool, not as a final publishing step.

The Validation Pipeline

Every piece of AI-generated text passes through an automated validation pipeline before it is eligible for publication. This pipeline checks:

Content that the pipeline flags as potentially problematic is routed for human editorial review. A human editor reads the flagged content, determines whether it contains a substantive error or policy violation, and either corrects it or clears it.

Our current substantive flagged-content rate — meaning content where the human reviewer confirmed a real problem requiring a correction, as opposed to a false-positive flag — is approximately 0.06% of generated outputs. Flagged content is not published until it has been reviewed and either corrected or cleared. Content confirmed to contain substantive errors is corrected before publication and the underlying prompt or pipeline configuration is reviewed to reduce recurrence.

What AI Does Not Do

AI is not used to make editorial judgements about which entities to feature, how to rank them, or what financial decisions readers should make. AI-generated content describes entities and explains concepts; it does not recommend, rank, or evaluate.

What We Do Not Do

We want to be direct about several things this site does not do, because they matter for how you should interpret what you read here.

Limitations You Should Know About

Data quality and timeliness have real limits, and we would rather be honest about them than overstate our reliability.

For any regulatory or legal purpose, you must go to the primary source.

Suggesting a Correction

If you believe a profile contains an error, outdated information, or misleading content, we want to know. You can reach our editorial team at corrections@luxembourgmedia.com. Please include the entity name, the specific field or statement you believe is incorrect, and — if possible — a reference to the source data you think should apply.

We review correction requests and, where a substantive error is confirmed, we update the content and note the correction. We aim to respond to correction requests within five business days.

For general enquiries, contact us at hello@luxembourgmedia.com. For data privacy matters, use privacy@luxembourgmedia.com.


Effective date: 31 May 2026. This methodology page is reviewed and updated when our data sources, processes, or editorial practices change in a material way.