micro1, an AI training company based in San Francisco, announced on 9 October 2026 that it plans to spend $1 billion over the next 12 months buying and licensing data from businesses. The data is meant to help teach AI agents how companies actually work. Everything below is micro1’s own account unless stated.
What micro1 announced
In its announcement, micro1 says it is “announcing a commitment to spend $1 billion over the next 12 months acquiring and licensing enterprise operational data” through its Company Data Partnerships programme [1]. It says the money comes from “capital provided by Citi and Hercules Capital” [1]. The post does not say how much has been spent or which companies have signed up. It is a statement of intent, not proof of money spent.
Why AI companies want this data
Operational data means the everyday material that shows how a business runs: procedures, internal documents, customer-relationship records, project histories and quality checks [2]. micro1 says it uses “de-identified company data to build reinforcement learning environments that reflect real business operations” [1].
In plain words, a reinforcement learning environment is a practice arena. An AI agent is given a task, tries it, and is scored on the result. micro1 says real business work involves “incomplete information, competing priorities and exceptions that require judgment”, and that training AI to cope needs environments that keep that context [1].
What micro1 says it will pay
micro1’s programme page says pay depends on “the quality, uniqueness, and value of the data” [2]. It lists three levels: $100k+ for a qualified partnership, $500k+ for large datasets or ongoing work across several teams, and $1M+ for “highly unique, proprietary operational data” [2]. TNW words these as “more than $100,000”, “more than $500,000” and “more than $1m” [3]. These are advertised starting points, not payments made.
Privacy and control, as micro1 states it
- Companies “retain ownership of their underlying data” [2].
- Personal information is removed “when relevant” [2]. That is not a promise that it is always removed.
- Some datasets may be rewritten synthetically, to keep the structure of the work while reducing links to the original records [2].
- Companies can review sample data before it is used, and micro1 says data is kept only for agreed periods [2].
These are micro1’s descriptions. TSN has not seen any contract terms.
Context from TNW
TNW reports that in September micro1 made a competing bid for Spirit Airlines’ customer records, and that the court-appointed privacy official in Spirit’s bankruptcy has since backed Google’s $10m deal for that data [3]. micro1’s own pages do not mention this.
The Bottom Line
micro1 says it will spend $1 billion over 12 months on company data, financed by Citi and Hercules Capital. Nothing independent confirms the spending or the payment levels yet. This is not investment advice.
Related on TSN: Mecka AI and Parallel Systems: Paying for Robot Data and Autonomous Rail
Sources
- micro1, “micro1 announces commitment to spend $1B on enterprise data acquisition over the next 12 months”, 9 October 2026 (company announcement, primary source). https://www.micro1.ai/blog/micro1-commits-1b-to-enterprise-data-acquisition
- micro1, “Monetize your company’s operational data” (Data Partnerships programme page, company source for payment levels, ownership, privacy and synthetic rewrites). https://www.micro1.ai/data-partnerships
- The Next Web, “micro1 commits $1bn to buy company data for training AI agents”, 9 October 2026 (source for the Spirit Airlines context and its wording of the payment levels). https://thenextweb.com/news/micro1-1bn-enterprise-data-ai-training-citi-hercules
