OpenAI tests visual ads during ChatGPT image generation

OpenAI announced on October 5, 2026, that it will test a visual ad format during image generation in ChatGPT, with the test planned to begin later in October in the United States. According to OpenAI, the ads will carry clear labels and remain separate from the generated image. OpenAI also announced expanded measurement and attribution tools for advertisers. The company’s announcement and a same-day report by Digiday confirm the date and the planned US test.

What OpenAI is testing

The new format is intended to let advertisers present products or services visually while a ChatGPT user is creating an image. OpenAI says the advertisement will not be incorporated into the image itself, and that advertising will not affect the answers ChatGPT provides. The test is not a general launch: OpenAI says it will start with an initial group of advertisers in the US.

The announcement also describes expanded campaign measurement. OpenAI says Hightouch, Tealium and LiveRamp integrations can help advertisers send conversion data from their existing systems to ChatGPT Ads. It lists attribution and reporting support from partners including AppsFlyer, Triple Whale, Adjust, Branch and Kochava, among others. OpenAI says work on geo-based incrementality experiments with Haus, Measured and WorkMagic is still at an early stage.

Why it matters to businesses that automate work

For businesses that automate marketing or back-office processes, the announcement matters less as a new image feature than as a potential new source of advertising activity and performance data. If the test expands, teams may need to handle another format, check how its results are recorded and bring those records together with website, app or sales data. That can create routine work in campaign reporting, reconciliation and follow-up, especially when data sits across separate platforms.

OpenAI’s expanded measurement options could help connect campaign information with systems advertisers already rely on. But availability of integrations does not, by itself, show that campaign attribution will be complete or comparable across tools. OpenAI describes its incrementality work as early-stage, so businesses should distinguish reported attribution from evidence that an ad caused an additional sale.

What changes in practice

For an operations team, the practical first step is not to redesign its workflow around an announced test. Instead, decide what evidence would be needed before committing time or budget: a clear campaign record, a defined conversion event, and a consistent way to compare the advertising platform’s report with the company’s own records.

These checks apply whether a business has one person managing operations or separate marketing, finance and data teams. Automation can reduce repeated copying and reconciliation, but it cannot make an unclear conversion definition reliable. Nor can it resolve conflicting attribution assumptions without a decision from the business.

What to watch next

Businesses should watch for confirmation that the test has started, details about advertiser eligibility and the format’s availability beyond the initial US group. They should also look for clearer reporting guidance, including how conversion data is matched and how results from different attribution partners should be interpreted. OpenAI’s October 5 announcement does not establish a broader rollout schedule.

For now, the announcement signals a planned test and expanded measurement infrastructure, not a proven advertising channel for every business. Any decision to participate should be based on actual access, the quality of the available reporting and whether the team can compare those results with its own business records.