For AI labs
European data, cleared for AI labs.
Real-world work data from European companies, refined in the EU and licensed with documented rights. Tell us what you're training, and we'll tell you what we can source.
Real work, with outcomes
Workflows and ERP records, support cases, code history, documents and agent traces: connected records that show how work moved from request to result.
Rights cleared per dataset
Legal basis, customer contracts, trade secrets and works-council involvement are reviewed and documented before anything is licensed.
Anonymised in the EU
Raw data is refined only in the EU. Anonymisation is assessed against EDPB guidance, and only the refined dataset is licensed.
Provenance for the AI Act
Each dataset comes with the source and rights documentation you need for your training-data summary and copyright policy.
Request data
Tell us what you're training
Share the capability and data you need. It takes two minutes, and everything except your contact details is optional.
- Can you source data to a specification?
- Yes. Tell us the capability, domain, language and volume you need. We source matching data from European companies and refine it into the format you train on.
- Which formats do you deliver?
- Refined datasets of complete work records, task collections with grading criteria, evaluation sets, and recurring refreshes as companies keep producing new data.
- Which countries and languages do you cover?
- We work with companies across Europe, of any size and industry, starting with German-speaking markets. Tell us which languages you need.
- Does any personal data leave the EU?
- No. Raw data is processed only in the EU, and what we license is anonymised, refined data that is no longer personal data under GDPR.