Data brokers are gathering companies' internal documents for AI training

single source· 1 articles · confidence: low · first seen 2026-11-06 16:00 UTC

What this means for you

Nothing to act on from what is published: no broker, buyer, price or contract term is named. The question worth asking internally is contractual — whether your supplier terms and employee data policies would let documents your organisation generates be resold as training data.

Data brokers are collecting internal corporate documents — files and emails produced inside companies for their own use — to sell as training material for AI models, according to Wired. The piece says the practice is drawing objections, but names no brokers, no buyers, no prices and no contract terms, so neither the size of the trade nor who is exposed can be established from it. It describes a market forming around material companies generate incidentally, not a product launch or a published measurement. No model, benchmark or vendor is named in what has been published.

Key facts

  • ·Data brokers are collecting internal corporate documents to supply as training data for AI models, Wired reports. source
  • ·The practice is drawing objections, though the article does not identify who is objecting. source
  • ·The report names no brokers, buyers, prices, volumes or contract terms. source
  • ·The report was published on 6 November 2026. source

What the sources say

  • Wired AI — Reports a market in internal company documents as model training material, and the discomfort it is causing.

Sources

The original reporting. Follow these — they did the work.

← the wire