OpenAI is introducing visual advertisements to ChatGPT alongside expanded measurement partnerships, moving to monetize its weekly base of 1.2 billion users beyond software subscriptions alone.
OpenAI will begin testing the visual ad format later in October 2026 in the United States with an initial group of advertisers. The format will appear during image generation for users on the Free and Go plans. Under the system’s design, the advertisements are clearly labeled and remain separate from the user-generated image. OpenAI stated that advertising will not influence the text answers or information ChatGPT returns to users.
To track commercial performance, OpenAI connected conversion data pipelines with Hightouch, Tealium, and LiveRamp. It also added reporting, Conversions API support, and click attribution integrations with providers including AppsFlyer, Triple Whale, Adjust, DV Rockerbox, Northbeam, Branch, Singular, Kochava, Airbridge, and Tenjin. Measurement partners reported early performance figures across initial trials: DV Rockerbox recorded that WeightWatchers achieved an attributed cost per acquisition 15.3 percent below its blended paid-search benchmark; WorkMagic documented a statistically significant lift for the wellness brand Dose, with 67 percent of incremental purchases coming from net-new customers; and Triple Whale tracked that 93 percent of Portland Leather’s visitors arriving from ChatGPT Ads were new to the site.
OpenAI noted that its incrementality measurement work remains in its early stages. To study causal outcomes rather than basic correlation, the platform is collaborating with Haus, Measured, and WorkMagic on geo-based testing frameworks. OpenAI also established partnerships with full-funnel and advanced measurement providers, including Fospha and INCRMNTAL.
Placement guardrails screen user interactions to keep advertisements away from emotionally vulnerable, sensitive, or unsuitable conversational contexts, combining automated filtering with human oversight and continuous monitoring. Qualifying advertisers can also use a feature called Negative Phrases to restrict placements according to internal brand guidelines. For third-party validation, OpenAI launched brand suitability pilots with DoubleVerify and Integral Ad Science, which test safety guardrails in controlled environments without accessing private user chats. DoubleVerify Chief Executive Mark Zagorski stated that context in ChatGPT is defined by the conversation rather than a static web page or video, requiring new mechanisms to align placement with suitability standards. Integral Ad Science CEO Lidiane Jones noted that IAS’s evaluation will examine how OpenAI’s safeguards operate across evolving scenarios while maintaining conversation privacy.
OpenAI opened registrations for advertisers through its dedicated portal at ads.openai.com. The company said conversations remain private and that users will retain control of their experience as it develops additional ad formats across ChatGPT.
