- ■
OpenAI charges $60 CPM for ChatGPT ads—triple Meta's typical rates—but won't provide conversion data advertisers get from Google and Meta
- ■
Early advertisers get only 'high-level' metrics: total impressions and clicks, not purchase attribution or user actions
- ■
For advertisers: You're paying premium prices without premium transparency—a sign the platform still lacks advertiser confidence signals
- ■
For investors: Watch whether advertisers actually take the deal; lack of conversion data transparency often predicts lower monetization success than incumbents
OpenAI is pricing ChatGPT ads at $60 per thousand views, triple what Meta charges, while deliberately withholding the conversion metrics that make ad platforms worth their premium. This isn't a pricing power demonstration—it's a signal of something more interesting: OpenAI's uncertainty about how to prove advertising value on an AI platform. The strategy reveals the gap between ChatGPT's user scale and the advertiser confidence needed to justify positioning against Google and Meta. For investors watching OpenAI's path to profitability, this matters less as a revenue signal than as a window into how the company balances monetization speed against platform maturity.
OpenAI just made a fascinating choice about how to monetize ChatGPT—and the choice reveals more about what OpenAI doesn't have than what it does. The company is asking advertisers to pay $60 for every thousand views of their ads in ChatGPT, according to The Information. That's roughly 3x what Meta typically charges for the same ad placement. But here's where the story shifts from pricing power into something less certain: OpenAI won't tell advertisers whether anyone actually acted on those ads.
When Google or Meta sell ads, they give buyers the full machinery of accountability. Conversion pixels. Attribution windows. Purchase data. Real-time dashboards showing which users clicked and bought. That transparency is why premium CPMs work—advertisers know what they're paying for. OpenAI is offering something different. Early advertisers on ChatGPT will get "high-level" data: total ad impressions, total clicks. Nothing more. No purchase data. No user actions beyond the click itself.
This isn't a power move. It's a gap made visible.
The context matters here. ChatGPT has roughly 200 million weekly active users, according to recent reports. That's real scale—enough scale that you'd think the platform could demand premium pricing. And OpenAI is demanding it. The $60 CPM positions ChatGPT's ads in the luxury tier alongside premium publisher placements. For context, Google's average CPM varies widely by industry, but premium placements often run $50-$100. Facebook typically runs $15-$20. So OpenAI is pricing like a premium publication, not like an emerging platform still proving advertiser ROI.
But pricing like a premium publication and being one are different things. The lack of conversion data suggests OpenAI knows this. If the platform's value proposition were ironclad—if advertisers were lining up to pay for access to ChatGPT users—the company could demand full transparency and still command premium rates. Instead, it's holding transparency back, possibly because early advertiser feedback showed hesitation about ChatGPT's actual conversion rates. Or it could signal OpenAI wants to avoid the complexity of detailed tracking while the ad platform is young. Either way, the gap between the ask and the transparency offered tells a story about confidence.
This mirrors a pattern Apple navigated when it entered digital advertising with App Store placements. Apple had massive scale and user trust, but it took time to prove advertiser ROI because the company heavily restricted tracking data. The difference: Apple had leverage from being the gatekeeper to the App Store. OpenAI needs advertisers to believe in the platform's value proposition, not just its reach. Withholding conversion data in that position is a gamble.
For advertisers, the calculus is brutal. You're paying 3x Meta's rates for an unproven audience and minimal transparency. That makes sense only if ChatGPT's user quality—intent, engagement, conversion rates—is genuinely superior to what you'd get elsewhere. Right now, that's an article of faith, not a proven fact. First-mover advertisers are essentially subsidizing OpenAI's data gathering phase. They'll run the campaigns, OpenAI will watch which ads drive engagement, and in six months the company will presumably publish "Here's how advertiser ROI looks on ChatGPT." By then, the pricing will likely come down.
The first ads on ChatGPT will start appearing in the coming weeks on free and Go plan users, with an important exception: users under 18 and conversations marked as sensitive (mental health, politics, etc.) won't see ads. That carve-out matters because it shrinks the addressable audience and excludes exactly the high-intent moments where advertising could prove most valuable. You're not running ads in financial conversations or health research—the moments where users might actually convert. You're running them in the remainder.
What's particularly interesting is the timing. OpenAI announced ads in ChatGPT earlier this month with privacy commitments: "We will never sell your data to advertisers." That's the right play for a platform built on trust. But it also constrains the advertiser value proposition. Without detailed user data, advertisers are flying somewhat blind. They know ChatGPT users are engaged—200 million weekly actives speaks to that. But they don't know if those users are the right kind of engaged for their products.
This is where Google and Meta built their empires. They don't just sell ads—they sell certainty. They've spent decades building the infrastructure to answer the advertiser's core question: "Will this audience buy from me?" Both platforms offer that certainty because they can track users across their networks. OpenAI is choosing a privacy-first model, which limits tracking, which limits certainty, which should logically limit pricing power. Instead, OpenAI is betting it can charge premium rates on brand trust and user intent alone.
That's not impossible. It's just harder than the incumbent playbook. And the willingness to withhold advertiser transparency suggests OpenAI knows it's a harder sell and is managing initial expectations accordingly.
OpenAI is doing something brave and risky: monetizing ChatGPT on brand and intent rather than data infrastructure. The $60 CPM pricing says one thing ("we're premium"), while the lack of conversion transparency says another ("we're still figuring this out"). For advertisers, this is a first-mover premium with first-mover risk. For investors, watch whether early campaign data proves ChatGPT users convert at rates that justify premium pricing. For enterprises considering where their ad budgets should flow, the question isn't whether OpenAI will eventually crack advertiser ROI—it will. The question is whether you want to be an early investor in that learning curve or wait for proven metrics. The timeline probably favors waiting.





