OpenAI has formed a dedicated advertising integrity team ahead of planned advertising tests inside ChatGPT, signaling that sponsored content is moving closer to the chatbot's interface.

The company has not yet announced launch dates, formats, pricing, or launch markets for the tests. The staffing move, however, confirms earlier reporting that OpenAI has been exploring an ads business for its consumer product rather than relying solely on subscriptions.

What does the new team actually do?

An ads integrity team typically handles the unglamorous but critical side of advertising: keeping fraudulent, misleading, or brand-unsafe ads out of the system. For a conversational product like ChatGPT, that job is harder than on a search results page.

Key questions the team will have to answer:

• How does an ad appear inside a conversational answer without degrading trust in the response?
• Which advertisers get access during the test phase, and under what vetting rules?
• How does OpenAI measure whether a sponsored recommendation influenced the model's output?

OpenAI has not publicly detailed the team's remit, headcount, or who leads it.

Who uses ChatGPT — and why advertisers care

ChatGPT is among the most widely used consumer AI applications globally, with hundreds of millions of weekly users according to OpenAI's own past statements. That scale makes it an obvious advertising inventory target, and an obvious integrity risk: a chatbot that mixes paid promotion into answers operates without the visual separation users get from sponsored links on a search page.

The move puts OpenAI on a collision course with Google's core business. Google already runs ads alongside AI-generated summaries in Search, and Microsoft monetizes Bing's Copilot integration. An ads-enabled ChatGPT would compete directly for the same brand budgets.

What it costs — and what we don't know yet

No pricing has been disclosed. OpenAI currently monetizes ChatGPT through subscription tiers such as ChatGPT Plus, and through API usage billed per token. Advertising would add a third revenue line, but the company has not said whether ads would target free users only, all users, or be opt-out for paying subscribers.

OpenAI has also not confirmed:

• the geographic scope of the initial tests;
• the ad formats under consideration;
• whether results will be published publicly.

The integrity problem is the story

Forming the integrity team before running the first tests is the notable sequencing here. Platforms that bolt on ad safety after launch tend to spend years chasing scams and misinformation. OpenAI appears to be attempting the reverse order: build the guardrails first, then sell the inventory.

Whether that discipline survives contact with revenue targets is the open question. The tests, when they arrive, will be the first hard evidence of how advertising and a conversational AI product can coexist.

We will update this story as OpenAI discloses test dates, markets, and format details.