ChatGPT Ads Pilot: $60 CPM, $200K Minimums, and Advertisers Already Pushing Back

OpenAI is six weeks into selling ads inside ChatGPT. The early numbers tell a conflicted story: a $60 CPM that’s roughly 3x Meta’s average, a $200,000 minimum buy-in for what’s essentially a beta test, and a rollout so slow that some of the biggest advertising agencies in the world are questioning whether they’ll get enough data to justify the spend.

The ChatGPT Ads Pilot launched on February 9, 2026, targeting Free and Go tier users in the United States. As of late March, ads have reached roughly 5% of ChatGPT’s mobile user base — up from just 1% at the start of the month. WPP, Omnicom, and Dentsu are all participating. Criteo signed on as the first ad tech partner, unlocking access for 17,000 advertisers. On paper, everything looks like a proper launch. In practice, the industry is already running into friction.

The Money Side: Premium Pricing, Limited Scale

The $60 CPM positions ChatGPT ads at the high end of digital advertising. For context, typical display advertising CPMs range from $2 to $10. Even premium search ads rarely touch $60. OpenAI is betting that conversational AI placements carry more intent — and early Criteo data supports this to a degree, with users referred from LLM platforms converting at approximately 1.5x the rate of other referral channels.

But that premium pricing comes with a catch: the $200,000 minimum commitment. For an alpha-stage test program, that’s roughly double what advertisers normally allocate to experimental placements. Some brands are spending $200K–$250K on a test that’s currently reaching 5% of mobile users. The math is hard to make work if you’re a CMO trying to justify ROI to a board.

Analysts project OpenAI will generate under $1 billion in ad revenue in 2026, scaling toward $25–30 billion by 2030. Those are big numbers, but they’re projections — and the current pilot is the foundation everything else depends on.

What Advertisers Are Actually Complaining About

Multiple reports from CNBC, Digiday, and Ad Age paint a consistent picture of advertiser frustration. The complaints fall into three buckets:

Not enough data. Early testers receive weekly CSV performance reports with impressions and click data. For enterprise advertisers accustomed to real-time dashboards, granular attribution, and conversion tracking, a weekly spreadsheet feels like going back to 2010. The lack of conversion tracking is a particular sore point — impressions and clicks provide directional signals but can’t tell an advertiser whether that $60 CPM actually drove a sale.

Targeting feels blunt. ChatGPT prioritizes conversational context over user demographics, which is a fundamental departure from how ad buying works on Meta, Google, or TikTok. Advertisers are used to slicing audiences by age, income, interests, and purchase behavior. ChatGPT offers topical relevance — your ad shows up when the conversation is relevant — but geographic and demographic targeting remains imprecise. For localized businesses, significant ad spend may reach users entirely outside their service area.

The rollout is too slow. OpenAI has been explicit that the conservative pace is intentional. “The goal right now is to learn and refine the experience for consumers before expanding it more broadly,” the company told CNBC. That’s a responsible approach from a user-experience standpoint, but it creates a tension: advertisers committed $200K+ expecting meaningful volume and insights. At 5% of mobile users, they’re not getting either.

The Criteo Angle: Making ChatGPT Ads Accessible

Criteo’s entrance as the first ad tech partner on March 2 is strategically significant. The pitch to its 17,000 advertisers is straightforward: go live with ChatGPT ads “in days” using your existing Criteo Performance Media setup. No new contracts, feeds, or tracking tags required.

Criteo’s pitch deck, reported by Digiday, frames ChatGPT as generating 70% of AI assistant traffic, making it the dominant surface for conversational advertising. The deck also cites that 63% of consumers use AI assistants for product discovery and research, and 53% for consideration-phase shopping.

This is Criteo’s bet that conversational commerce is the next major ad channel — compressing the discovery-to-purchase journey into a single chat session. Whether that thesis holds depends entirely on whether OpenAI can deliver the measurement infrastructure that performance advertisers demand.

How ChatGPT Ads Stack Up Against Competitors

The conversational AI advertising space is still in its earliest days, but competitive dynamics are already emerging.

Google runs ads within AI Overviews and AI Mode in Search but has kept Gemini itself ad-free. This positions Google to capture AI-adjacent ad revenue without risking the user experience of its standalone AI product.

Perplexity tested sponsored placements throughout 2024 and early 2025 — then pulled back entirely in February 2026. Perplexity’s leadership concluded that even clearly labeled ads risk eroding user trust in an answer engine. That’s a direct philosophical counter to OpenAI’s approach.

Anthropic has positioned Claude as explicitly ad-free, competing on privacy and user experience as differentiators.

OpenAI is the only major AI lab pushing aggressively into advertising within its core conversational product. That’s either visionary or risky, depending on how the next few quarters play out.

User Sentiment: The Backlash Is Real

The user reaction has been overwhelmingly negative. According to community tracking, 68% of Reddit commenters expressed negative sentiment toward the announcement, and discussions on X drew over 10 million views, mostly skeptical.

The sharpest criticism comes from two angles. First, Go tier subscribers pay $8/month and still see ads — breaking the implicit contract that paying customers shouldn’t be monetized through advertising. Second, power users and early adopters feel this is a bait-and-switch: OpenAI built its user base on a clean, ad-free product, then introduced monetization once the audience was captive.

Privacy concerns add another layer. OpenAI’s updated privacy policy allows it to receive purchase data from advertisers when users click ads and make purchases. U.S. Senator Ed Markey sent letters to OpenAI and other AI companies expressing concerns about advertising in conversational AI platforms, citing risks around manipulation and the need for stronger safeguards — particularly for vulnerable users seeking health information.

OpenAI maintains that ads never influence ChatGPT’s responses, are clearly labeled as sponsored, and that advertisers don’t access individual chat content. Whether those guardrails hold as the program scales is the open question.

What Happens Next

The pilot runs through the end of March. What comes after depends on whether OpenAI can solve the measurement problem. Advertisers don’t need ChatGPT ads to be cheap — they need them to be provably effective. Right now, the data infrastructure isn’t there.

OpenAI also has to decide how fast to scale. Too slow, and advertisers won’t renew. Too fast, and user backlash could accelerate migration to ad-free alternatives like Claude or a repositioned Perplexity. The company is walking a tightrope between Wall Street-friendly revenue projections and the trust of 400 million weekly users.

The first wave of brands — Target, Adobe, Audible, Ford, Mazda, Williams-Sonoma — are watching their numbers closely. If the 1.5x conversion rate holds at scale and OpenAI delivers better attribution tools, the $25 billion projection starts looking plausible. If not, the ChatGPT Ads Pilot becomes a cautionary tale about pricing before product-market fit.

FAQ

How much does it cost to advertise on ChatGPT?
The ChatGPT Ads Pilot operates at a $60 CPM (cost per thousand impressions) with a minimum commitment of $200,000 to participate in the beta test. This pricing is significantly higher than typical display advertising ($2–$10 CPM) and roughly 3x the average cost of Meta ads.

Who can see ChatGPT ads?
Currently, ads appear only for U.S.-based users on ChatGPT’s Free and Go ($8/month) subscription tiers. Users on Plus ($20/month), Pro ($200/month), Business, Enterprise, and Education plans do not see ads. The rollout has reached approximately 5% of ChatGPT mobile users as of March 2026.

How do ChatGPT ads compare to Google and Perplexity ads?
Google places ads in AI Overviews within Search but keeps Gemini ad-free. Perplexity tested sponsored placements but abandoned the approach entirely in February 2026, concluding ads erode user trust. Anthropic’s Claude remains explicitly ad-free. OpenAI is currently the only major AI lab running ads inside its core conversational product.

Do ChatGPT ads affect the AI’s responses?
According to OpenAI, ads do not influence ChatGPT’s answers. Responses are optimized for helpfulness, and ads are visually separated and labeled as sponsored. Advertisers receive only aggregate performance data and do not have access to individual user conversations.

What brands are advertising on ChatGPT?
The first wave of advertisers includes Target, Adobe, Audible, Williams-Sonoma, Ford, and Mazda. Over 30 Omnicom Media clients have secured placements, and Criteo’s partnership opens access to 17,000 additional advertisers across categories including retail, automotive, beauty, CPG, and hospitality.

Top AI Product

Leave a comment Cancel reply