# Sell video game studio data to AI companies

> Game studios produce some of the richest working records in any industry – design documents, production chat, task tracking, builds and art, all tied to one shipped product. AI companies want that record today. You keep ownership, and publisher contracts and licensed third-party assets are assessed first.

By [Linus Näslund](https://corpusmarket.com/about.md), Co-Founder

## Why AI companies want game development data right now

AI labs are training agents that write code, generate assets, plan projects and coordinate teams. Game development combines all four. A single title involves designers, programmers, artists, audio, QA and producers working against milestones for years, and every decision leaves a trace – a design doc revision, a Jira ticket, a heated Slack thread about scope, a build that broke and got fixed.

Very little of that is public. Shipped games, trailers and the occasional postmortem show the end result. The months of iteration, playtest feedback, cut features and production trade-offs stay inside the studio. For AI companies building tools for software and creative work, that complete record is hard to find anywhere else, which makes game studios a good fit for AI data licensing today.

## What carries the value

| Area | What it shows | Typical systems |
|---|---|---|
| Design documentation | Game design docs, feature specs, level and economy design, and how they changed | Confluence, Notion, Google Drive, SharePoint |
| Production chat | Daily coordination, scope decisions, bug triage and crunch-time problem solving | Slack, Discord (internal), Microsoft Teams |
| Task tracking | Tickets, sprints, bug reports, estimates and how work was prioritized | Jira, Hansoft, Shotgrid, Linear |
| Builds, code and assets | Source the studio owns, version history, art and audio with clear ownership | Perforce, Git, Plastic SCM |
| Email, CRM and finance | Publisher and partner communication, business development, budgets and transactions | Email, HubSpot or Salesforce, QuickBooks, Xero, NetSuite |

The links between these are what make the record valuable. A Jira ticket tied to the design doc it implements, the chat thread where the approach was argued and the commit that closed it shows how software and content actually get made. Code repositories can sometimes stand on their own, but for most studios the full record across systems is what drives the offer, and leaving whole systems out lowers it.

## Typical starting size

Bigger teams get bigger offers. Count the people who work in the studio's systems day to day – designers, engineers, artists, QA, producers and the business side. Contractors and outsourcing partners who only delivered files add less than people who were in the chat and the tracker.

Years matter: at least three years of records still in the systems, ideally across more than one title, and five or more years is better. Internal communication should mainly be in English, which can be a question for studios with teams in several countries.

The [self-check guide](https://corpusmarket.com/guides/does-my-company-data-qualify.md) covers each point, and the [industry overview](https://corpusmarket.com/guides/which-industries-have-the-most-valuable-data.md) shows how studios compare with other sectors. Studios that also run a platform or live service may find [tech company data](https://corpusmarket.com/sell-tech-company-data.md) relevant too.

**[See what your data is worth](https://corpusmarket.com/submit-data)** – Free estimate in two minutes – not an offer

## What is assessed first, and what is never part of a deal

- **Publisher contracts.** Funding and publishing agreements often assign IP or restrict use of code, assets and confidential information. Each title is assessed against its contract.
- **Licensed IP.** Games built on a licensed brand, film or sports property usually carry strict terms from the licensor and are normally left out.
- **Third-party engines, middleware and assets.** Engine source, plugins, marketplace assets, fonts, licensed music and stock audio are filtered out.
- **Player data.** Player accounts, telemetry tied to individuals, payment records and support tickets with personal information are not part of the development record and are excluded or de-identified.
- **HR and personal data.** Reviews, hiring and personal conversations are filtered inside email and chat, and names and identifiers are removed before delivery.

This describes how it generally works today, not legal advice. Every studio and every deal is unique, and what can be included is assessed case by case.

## What companies typically receive right now

| Full-time US staff | Indicative range |
|---|---|
| 20–49 | $200K–$500K |
| 50–149 | $400K–$900K |
| 150+ | $800K and up |

Smaller companies in high-demand industries are quoted individually. These are indicative ranges from what we see in the market right now, not offers – the real number comes in a written offer after a review, and depends on your industry, years of history, which systems are included and demand at the time. Our [data value estimator](https://corpusmarket.com/what-is-your-data-worth.md) gives a first estimate in about two minutes.

## How a deal works

1. **You describe the studio** – team size, years and titles covered, and the tools you use. No files are uploaded.
2. **We assess** ownership, publisher and licensor terms, and demand, and get back to you within 48 hours on business days.
3. **You approve a written offer** – or say no. Nothing is shared before you approve.
4. **Filtering and delivery.** Third-party, restricted and personal material is filtered out inside each system, identifiers are removed, and the agreed record is delivered.

The license is non-exclusive by default and limited in scope, time and use, and the studio keeps its IP. A review costs nothing – we're paid only when a deal goes through. More on [how it works](https://corpusmarket.com/how-it-works.md). For the full picture across industries, see how to [sell data to AI companies](https://corpusmarket.com/sell-data-to-ai-companies.md).

## Frequently asked questions

### What game studio data do AI companies want?

The development record – game design documents, production chat, Jira or similar task history, builds, code and assets the studio clearly owns – together with email, file storage, CRM and accounting. The value is in how a game was planned, built, tested and shipped, not just the final release.

### What if a publisher funded the game?

Publisher agreements often decide who owns the IP, code and assets, and many include confidentiality terms. Each title is assessed against its contract, and material the studio doesn't own or can't share is left out.

### Can licensed engines, middleware or asset-store content be included?

Usually not. Engine source, middleware, plugins, fonts, music and marketplace assets come under their own licenses and are filtered out. The studio's own code, design and art are what carry the value.

### Does the studio give up its IP?

No. A deal is a training license, non-exclusive by default and limited in scope, time and use. The studio keeps ownership of its games, code and assets.