OpenAI announced on October 5 that it’ll begin testing a visual ad format inside ChatGPT’s image-generation surface, placing labeled product imagery alongside pictures users generate themselves. The pilot is U.S.-only, limited to Free and Go plan users, and seeded with a closed initial roster of advertisers. The newsworthy part isn’t the format. It’s the measurement stack that arrived with it.
On day one, OpenAI is wiring in click-attribution partners that read like a complete DTC marketing org chart: AppsFlyer, Triple Whale, Adjust, DV Rockerbox, Northbeam, Branch, Singular, Kochava, Airbridge and Tenjin, per TechCrunch. Full-funnel work goes to Fospha, Measured and INCRMNTAL. Conversion data flows through Hightouch, Tealium and LiveRamp, according to BleepingComputer. Incrementality is handled by Haus, Measured and WorkMagic, per OpenAI. DoubleVerify and Integral Ad Science are running a brand-suitability pilot. For a format that doesn’t formally exist in the wild yet, the plumbing is unusually finished.
That matters because of what OpenAI is quietly telling advertisers it needs the format to do. The company has guided investors to $2.5 billion in ad revenue this year and a $100 billion target by 2030, with ChatGPT ads hitting a $1 billion annualized run rate at the end of August, Digiday reports. ChatGPT reaches 1.2 billion people each week. A measurement partner list this exhaustive is how you convert that reach into performance budgets rather than brand experiments.
The open question is whether the surface itself converts. Donatas Smailys, CEO of creator marketing firm Billo, offered the skeptics’ case bluntly: “Search ads work because they meet people at the moment they’re looking for something. Someone generating an image in ChatGPT is not shopping.” OpenAI’s bet is that visual intent is its own signal, closer to a Pinterest board than a Google query. Samantha Jacobson, OpenAI’s vp of monetization partnerships, told an Advertising Week panel this week that “knowing who the person is, I think matters less.”
That framing is a convenient one, because OpenAI hasn’t detailed the targeting mechanics or whether creatives will be pulled from a submitted library, as The Register notes. It also lands days after ChatGPT Ads added CRM targeting with a 25,000-user floor that priced out most small advertisers, and days after OpenAI’s sponsored agents format and the sponsored-agents interview format introduced two more surfaces in the same month.
Four ad products in two weeks, each with its own targeting logic, each with the same measurement partners quietly stapled on. The pilot is a trial balloon. The attribution roster is the actual announcement.
Sources
- https://openai.com/index/new-chatgpt-ads-format-and-measurement/
- https://techcrunch.com/2026/10/05/openai-launches-visual-ads-that-appear-alongside-image-generation-results/
- https://digiday.com/media-buying/openai-would-like-to-remind-marketers-that-its-ads-are-for-the-good-of-humanity/
- https://www.theregister.com/ai-and-ml/2026/10/05/openai-to-flood-your-eyeballs-with-visual-ads/5301152
- https://www.bleepingcomputer.com/news/artificial-intelligence/openai-will-show-visual-ads-in-chatgpt-while-you-generate-images/