On October 5, 2026, OpenAI announced it will test a new visual ad format in ChatGPT, starting later in October in the United States, and is expanding measurement and attribution integrations for advertisers.
What happened
OpenAI introduced a visual format that shows product inspiration, product usage or the experiences a product enables, to be tested first during image generation in ChatGPT. The initial test targets users on the Free and Go plans. Ads are labeled and kept separate from the generated image, and the company states that advertising does not influence ChatGPT’s answers. Businesses can sign up at ads.openai.com; the exact start date is not specified beyond «later in October».
Key facts
- OpenAI cites 1.2 billion weekly ChatGPT users.
- Conversion data integrations: Hightouch, Tealium and LiveRamp.
- Web and app attribution with partners including AppsFlyer, Triple Whale, Adjust, DV Rockerbox, Northbeam, Branch, Singular, Kochava, Airbridge and Tenjin.
- Incrementality tests, at an early stage and based on geo experiments, with Haus, Measured and WorkMagic.
- Brand suitability pilots with DoubleVerify and Integral Ad Science, without access to private conversations.
Partner results cited by OpenAI: WeightWatchers’ attributed cost per acquisition was 15.3% lower than its blended paid-search benchmark (per DV Rockerbox); WorkMagic reported a statistically significant lift for wellness brand Dose, with 67% of incremental purchases from net-new customers; and Triple Whale reported that 93% of Portland Leather’s ChatGPT Ads visitors were new. These are individual cases reported by the platform and its partners.
Context
Kantar’s Media Reactions 2026 study, published October 2 with more than 800 senior marketers and 23,000 consumers, captures the tension: a net 75% of marketers plan to increase spending on AI assistants in 2027, yet only 32% of consumers use them to research brands and 23% trust their brand recommendations.
Strategic takeaway — Studio 24-7
ChatGPT is emerging as a new ad channel, but still in testing and with immature measurement. The smart move is neither to rush in and spend nor to ignore it: it is to prepare the system. That means having measurement solved (conversion events, attribution, CRM), defining a capped test budget and comparing against your current channels on a common metric, such as cost per acquired customer.
Systems before improvisation: whoever already measures well can test a new channel in weeks and learn fast; whoever does not measure just spends. Results before appearance.
Sources
- OpenAI: New ChatGPT ads format and measurement — October 5, 2026
- Kantar: Media Reactions 2026 — October 2, 2026