Amazon has struck a deal with OpenAI that lets advertisers buy ads in ChatGPT through Amazon’s advertising platform. As a result, Amazon becomes another door into ChatGPT’s audience for its advertising customers, and OpenAI gets access to Amazon’s advertiser base, sales infrastructure, and ad technology.
On the surface, Amazon Ads has little reason to negotiate a deal with ChatGPT. Amazon Ads is a $70 billion-a-year business, the third-largest digital ad platform after Google and Meta. ChatGPT Ads, by comparison, was running at roughly $1 billion annualized when this deal was announced. More importantly, AI assistants compete with Amazon for product discovery. Someone shopping for a retailer might now start that research in ChatGPT instead of on Amazon.
But this deal is not really about ChatGPT specifically. It’s the latest move in a pattern Amazon has followed for close to two years: locking up premium ad inventory it doesn’t own. Amazon DSP has added Netflix, Roku, Spotify, Disney, and ESPN as ad supply over that stretch. And now, ChatGPT. Amazon’s pitch to advertisers is: buy this inventory through Amazon and get better targeting and lower fees than buying it directly or through a rival platform. According to Digiday, Amazon’s fees on comparable guaranteed ad deals run close to zero, versus a historical 15% to 20% cut for The Trade Desk. If that pattern holds for ChatGPT inventory too, it would give advertisers a reason to buy through Amazon rather than dealing with OpenAI directly, though no source has confirmed the fee structure on this specific deal.
Why the Deal Matters
The deal moves ChatGPT advertising into the mainstream media-buying market. An advertiser already running campaigns through Amazon DSP doesn’t need a new buying relationship or a separate campaign operation to test ChatGPT. It also fast tracks OpenAI into the advertising market. OpenAI doesn’t have to recreate advertiser relationships, buying technology, or measurement infrastructure from nothing. It can plug into a company that already has all three.
For Amazon, the prize is bigger than ChatGPT. Google and Microsoft are both building their own advertising models for Gemini and Copilot. If Amazon becomes the default way advertisers buy inside conversational AI generally, not just on ChatGPT, it extends its ad dominance well past retail and into whatever comes next in AI-driven search and shopping.
OpenAI has been building its own commerce and checkout capabilities, including partnerships that let people complete purchases without leaving ChatGPT. That’s a possible threat to Amazon’s retail funnel if it matures. Striking an advertising deal now gives Amazon a presence inside ChatGPT while that potential threat is still developing, rather than waiting to be cut out of a shopping experience it doesn’t control.
What the Deal Means for Advertisers
For advertisers, the biggest implication is that conversational AI is becoming easier to fold into existing media plans. That creates opportunities and open questions, such as:
- ChatGPT becomes easier to test. An advertiser already working through Amazon can experiment without setting up an entirely new advertising operation.
- The context is different from conventional digital advertising. Someone comparing televisions or researching a business purchase is actively expressing a need, and the conversation gives OpenAI information about what that person is trying to accomplish.
- ChatGPT ads may influence a purchase without producing an immediate click, which makes their impact harder to measure. Someone could encounter a brand in ChatGPT, keep researching elsewhere, and buy days later. A last-click model won’t capture that influence.
Advertisers shouldn’t assume an ad built for another channel belongs in ChatGPT. People use an AI assistant to learn, decide, compare, or accomplish something specific, and advertising has to make sense inside that experience, or it won’t land.
What Advertisers Should Do Next
Advertisers don’t need to overhaul their media plans because Amazon and OpenAI struck a deal. They should start learning where conversational AI actually belongs within those plans.
- Test before reallocating significant budget. Use controlled spending and compare ChatGPT against channels you already buy. Easier access through Amazon doesn’t prove ChatGPT advertising will perform better.
- Choose situations where conversation adds something. Products involving substantial research and consideration are a reasonable place to start. Ask why reaching someone mid-conversation would beat reaching that same person somewhere else.
- Develop creative built for conversational environments. Don’t repurpose a banner or search ad without changing anything. Advertising here should provide information relevant to what someone is trying to accomplish.
- Measure more than clicks. Look at assisted conversions, branded search, site visits, and eventual sales where your measurement allows it. Determine whether conversational advertising changes decisions even without an immediate conversion.
- Compare AI platforms individually. Don’t build one “AI advertising” strategy and apply it everywhere. ChatGPT, Gemini, Copilot, Perplexity, and whatever follows will likely develop different audiences, formats, buying models, and consumer behavior. Advertisers should figure out which platforms actually matter to their customers.
Bottom line for advertisers: does reaching someone while they’re actively discussing a need or researching a decision produce results you aren’t already getting from the media you buy today? If it does, increase the investment. If it doesn’t, don’t spend money there just because conversational AI is getting attention.
True Interactive can help you navigate the advertising landscape, ranging from Google to ChatGPT. Contact us to learn more.
