What you keep
Your records, your business, and ownership of the data. A licence grants specified rights to use it, never ownership.
The marketplace for enterprise data.
Recent AI data deals
Reported deals involving large platforms. They show the market is real; they do not set a price for any one archive.
Almost none of the world's data is public. The rest is private. That's the supply AI labs can't reach.
Proportions shown are illustrative, not measured.
We connect businesses with companies willing to pay to license the data, while you retain ownership.
Based on published rates for expert-produced AI training data.
Replacement cost: what a lab would pay to commission this work from scratch, discounted because an existing archive is messier than data built to order.
Assumes 1,600 working hours per person per year.
Illustrative estimate in US dollars. Not an offer, not a valuation, and no sale or payment is guaranteed. Real value depends on what your records contain, who wants them, and what rights you can grant.
From codebases and design records to the private data already inside your company.
Get a detailed assessmentDrawings, models, revisions and tolerances, traceable to something that was actually made.
Do not send raw records during an assessment. We start with counts and descriptions; data moves only after a deal is agreed.
You have no real benchmark for price or terms.
We compare offers and help negotiate the best deal.
Illustrative example, not a real deal. Actual offers depend on what your records contain, who wants them, and what rights you can grant.
List the systems you use and roughly what's in them. You don't upload anything or hand over logins. We'll tell you if it's worth licensing and give you a ballpark figure.
You choose what goes in and what stays out. We get the rights signed off in writing and replace names and contact details. Then we package it so buyers can see what they'd be paying for.
We take it to AI companies, put their offers side by side and push for the best price. Your data doesn't go anywhere until you sign, and you're paid on the terms in that contract.
We swap names, companies and contact details for consistent labels and take out anything sensitive. The work itself stays intact, because that's the part buyers pay for.
An example from a work email
Dave Mitchell signed off the revised bracket drawing for Hartley Engineering. Send the updated tolerance sheet to d.mitchell@hartleyeng.com.au.
Dave Mitchell confirmed the part passed inspection on the second run.
The same email once we've prepared it
Person 07 signed off the revised bracket drawing for Company 03. Send the updated tolerance sheet to Email 07.
Person 07 confirmed the part passed inspection on the second run.
Dave Mitchell becomes Person 07 everywhere in the set, so a thread still follows one person. Companies and email addresses get labels too.
Passwords, payment details and anything you've ruled out come out before a buyer sees a single record.
The problem, the decision and the outcome stay joined up. That's the part an AI model learns from.
Made-up example. Replacing names is one step, not a guarantee: a record can still point to someone through context, so we review for that as well, and the limits go in the agreement.
Before you decide
Your records, your business, and ownership of the data. A licence grants specified rights to use it, never ownership.
Credentials, payment details, and anything you exclude. You set the scope before anything moves.
Permitted uses, restrictions and excluded buyers, written into the agreement rather than agreed in a call.
Buyers lined up and ready for supply, including several from the AI labs. We know what they are looking for.
Straight answers
Nothing is uploaded to start with. These are the things worth checking before you commit to anything.
It depends on what the records contain, how much there is, whether you hold the rights, and what buyers are looking for at the time. The estimator on this page is an illustration built from published rates, not an offer. An assessment gives you a better-informed answer, still not a guaranteed figure.
Yes. We have buyers already lined up, including several from the AI labs, and we know what they are looking for.
No. You can exclude systems, folders, particular clients or whole workflows. The agreed scope defines exactly which records are offered, and it is set before anything moves.
Yes. A licence grants specified rights to use the data for agreed purposes. It is not a sale of the data and not a transfer of ownership. You do need the right to license it in the first place, including any permissions owed to clients or third parties, and that is one of the first things we check.
Not on its own. A record can still identify someone through context, especially in a small industry where the job itself is recognisable. Replacing names is one step. The review has to cover identifying context and sensitive fields as well, and the limits belong in the agreement rather than in a promise.
Yes, and this is the moment to raise it. Competitors, specific clients, or uses you are not comfortable with can be excluded in writing. Restrictions agreed in conversation and not written down are worth nothing later.
Do not send raw records, credentials, API keys or payment details through this website or by email. Start by telling us which systems you use and roughly what is in them. Data moves only after a deal is agreed, and into the environment named in that agreement.
The assessment is a conversation, not an export. Beyond that it depends on how your records are stored and how many systems are involved, so we would rather scope it with you than quote a figure here that turns out to be wrong.