Updated October 2026
Can you sell your company's data to OpenAI or Anthropic?
What labs publish
The deals labs have published are with publishers
OpenAI's published content deals are with News Corp (reported at more than $250M over 5 years), the Financial Times ($5M-$10M a year) and Axel Springer (about $13M a year). Google pays Reddit about $60M a year, Reuters reported. We found no published program from OpenAI or Anthropic that buys operating data directly from businesses.1,2,4,5
There is one notable exception to the publisher pattern: in Spirit Airlines' bankruptcy, Google bid $10M directly for the company's internal archive.6
| Fact | Value and source |
|---|---|
| News Corp and OpenAI | More than $250M over 5 years (WSJ) |
| Financial Times and OpenAI | $5M-$10M a year |
| Axel Springer and OpenAI | About $13M a year |
| Reddit and Google | About $60M a year (Reuters, citing sources) |
| Spirit Airlines internal archive | $10M from Google, against a $7.5M Mercor bid |
How companies reach the labs
Through licensors that buy company data
micro1, Mercor, Polyshares, Nyne, Cognor Labs, Ventra and Troveo publish programs to license operating data from companies. micro1 says it licenses SOPs, knowledge bases, CRM metadata and project history, and has committed $1B over 12 months to buying company data. MIT Technology Review reported that AI labs are buying training data from startups that have shut down.7,8,9,10,3,11
These licensors generally do not say which labs receive the data. That is the first question to ask: which AI labs will receive it, and will you tell us?
What they want
What labs' suppliers ask companies for
Mercor lists Slack, Teams, Gmail, Outlook, Confluence, Salesforce, GitHub, Drive, Notion and 100+ other tools. micro1 asks for internal docs, SOPs and playbooks, knowledge bases, project workflows, internal comms, templates, CRM and workflow metadata, and task histories.12,13
Size floors apply. micro1 asks for 30+ employees, US first and mostly English data. Nyne asks for 20+ US W-2 employees and 3+ years in business. Polyshares says any company with real operating history.14,15,16
FAQ
Questions, answered with sources
We found no published Anthropic program that buys operating data from businesses. Companies that want to license to the labs work through licensors; micro1 has committed $1B over 12 months to buying company data.3
Most do not publish it. Ask before you sign, and ask whether the license lets them sublicense to anyone else. Troveo, for one, describes its deals as licensing on a revenue share.9
We do not answer legal questions. Ask the licensing company: their team reviews this with you before anything is signed.
Before you sign
Legal questions go to the licensing company
We do not answer legal questions. Ask the licensing company before you sign: their team reviews these with you.
Questions for this kind of business
- Do customer, vendor or platform contracts, NDAs or API terms restrict this?
Questions for any buyer
- Is the license exclusive, and for how long?
- Can you sublicense the data, and to whom?
- Which AI labs will receive it, and will you tell us?
- How long do you keep it, and how is it deleted?
- What is your liability cap, and who indemnifies whom?
- Will the contract ban re-identification of people in the data?
- Who removes personal data, by what method, and do we approve it first?