OpenAI announced a new visual advertising format and expanded ad measurement for ChatGPT on October 5, 2026. The first test will begin later this month in the United States with an initial group of advertisers, placing ads during image generation. The announcement is not a rollout to every country or every type of ChatGPT conversation. Source
What users may actually see
The format uses images to show product inspiration, usage or related experiences. OpenAI says each ad will be clearly labeled and kept separate from the image a user is creating. It also says advertising will not influence ChatGPT’s answers. In practical terms, an eligible user may see an ad while generating an image, but the announcement does not say the ad becomes part of the generated image or a basis for the assistant’s answer. Source
OpenAI did not give an exact launch day, the share of US users in the test or an ad-frequency limit. It also did not announce when the format will reach users outside the US or ordinary text conversations. Users should check the live interface for clear labeling and separation once the test begins.
How advertiser measurement is expanding
Advertisers can use Hightouch, Tealium and LiveRamp integrations to send conversion data from existing systems to ChatGPT Ads. OpenAI also lists web and app measurement partners, including AppsFlyer, Adjust and Branch, for conversion APIs, reporting and click attribution. Early work with Haus, Measured and WorkMagic is exploring geo-based experiments for incrementality measurement. Source
Attribution connects a purchase or signup to an advertising touchpoint. Incrementality asks whether the campaign produced additional outcomes compared with what would probably have happened without it. The cost and new-customer figures in the announcement are early findings reported by measurement partners. They are not independent OYOPICK tests or a guarantee that other advertisers will reproduce the same results. Source
What the privacy and suitability claims cover
OpenAI says placement guardrails assess whether a conversation is suitable for advertising and are intended to keep ads out of emotionally vulnerable, sensitive or otherwise unsuitable contexts. It also describes automated review, human oversight and ongoing monitoring. Qualifying advertisers can use Negative Phrases for narrower placement exclusions. Source
DoubleVerify and Integral Ad Science are joining controlled brand-suitability evaluation pilots. OpenAI says those evaluations will not give the partners access to private user conversations. The announcement is not a complete help page for user-level ad selection, data retention or every control available in the interface. Users should review the live ad and privacy settings when the test appears, while advertisers need to confirm eligibility, supported integrations and measurement terms after signing up. Source