ChatGPT Crossed $100M in Ad Revenue in 60 Days. The CPM Fell From $60 to $25. Here's the Business Model.
OpenAI's advertising pilot hit $100M ARR in under two months. The CPM has already halved. Inside the pricing collapse, the CPC pivot, and whether the $2.5B 2026 target is achievable.
By Marcus Johnson, Brand & Culture · May 23, 2026
ChatGPT hit $100M in ads in 60 days. CPM fell from $60 to $25. Inside OpenAI's advertising model, the CPC pivot, and what the $2.5B revenue target means.
Frequently Asked Questions
How do ChatGPT ads work?
ChatGPT ads appear as clearly labeled sponsored placements within the ChatGPT interface when users ask questions in commercial intent categories — software recommendations, travel, financial services, consumer products. The format is conversational rather than banner-style: a sponsored response or recommendation appears alongside or within the AI's organic answer. Targeting is intent-based, not behavioral. OpenAI does not use third-party data or tracking pixels. Instead, advertisers select categories that match their product, and placements appear when users ask questions in those categories. The result is a format that more closely resembles keyword-level Google Search targeting than Meta's audience targeting, but delivered in the context of an AI conversation the user has already trusted to give them a useful answer. Ads are available through both a managed buying process with a $50,000 minimum and a self-serve Ads Manager platform that opened to all US advertisers on May 5, 2026.
What is the CPM for ChatGPT ads in 2026?
ChatGPT ads launched in February 2026 at a $60 CPM (cost per thousand impressions), with a $200,000 minimum spend requirement. That opening CPM was among the highest in digital advertising history, reflecting the novelty premium and high intent quality of the format. By mid-April 2026 — approximately ten weeks after launch — clearing CPMs had compressed to approximately $25 as OpenAI expanded the advertiser base and opened more query categories to monetization. The $25 CPM remains a significant premium over Google Search ($8-$15 CPM equivalent) and social media advertising ($7-$12 CPM), but the compression from $60 to $25 reflects standard new-inventory pricing dynamics: early adopters pay for novelty and scarcity, then the market finds a sustainable clearing price as supply expands. OpenAI simultaneously introduced CPC (cost-per-click) bidding at $3-$5 per click to attract performance advertisers who prefer outcome-based pricing.
How much revenue is OpenAI making from advertising?
OpenAI's ChatGPT advertising pilot crossed $100 million in annualized recurring revenue (ARR) in under two months of operation, with approximately 600 advertisers participating. The company has set a 2026 advertising revenue target of $2.5 billion — roughly 25× the annualized rate at the two-month mark — which requires either dramatically expanding the advertiser base, significantly increasing average spend per advertiser, or both. OpenAI has also stated longer-term targets of $11 billion in advertising revenue by 2027 and $100 billion by 2030. The 2030 figure would make OpenAI's ad business comparable in scale to Meta's current advertising revenue. The 2026 target of $2.5 billion is aggressive but directionally plausible given ChatGPT's 800 million weekly users and the relatively high intent quality of conversational AI queries. Whether actual 2026 performance matches the target depends on advertiser adoption rates, CPM stability, and demonstrated conversion performance across categories.
Are ChatGPT ads better than Google ads for brands?
ChatGPT ads and Google Search ads serve similar intent-stage audiences — users actively seeking answers or recommendations — but differ meaningfully in several dimensions. ChatGPT ads operate in a higher-trust environment: users have already received an AI-generated answer they found useful, and a relevant sponsored suggestion may inherit some of that credibility. Google Search ads operate in a more skeptical environment where most users have learned to scroll past sponsored results. The tradeoff is that ChatGPT's conversion data is nascent — advertisers have no multi-year performance benchmarks the way they do for Google. ChatGPT also has a narrower commercial intent inventory than Google's billions of daily searches, though this is changing as the platform scales. For brands evaluating early-mover participation, the practical approach is to allocate a test budget alongside existing Google spend to generate comparable performance data, rather than shifting budget away from proven channels before ChatGPT benchmarks are established.
What is the minimum spend to run ChatGPT ads?
As of May 2026, ChatGPT ads are available through two buying mechanisms. The managed buying program — working directly with OpenAI's ad team — has a $50,000 minimum spend requirement, reduced from the $200,000-$250,000 minimum at launch. The self-serve Ads Manager, which opened to all US advertisers on May 5, 2026, has no stated minimum spend requirement for self-serve campaigns, though the platform is still in beta and access is limited. Advertisers on the self-serve platform can use either CPM bidding (default max bid of $60, with average clearing rates around $25) or CPC bidding at $3-$5 per click. The reduction of the minimum spend threshold and the launch of self-serve access represent a deliberate scaling strategy: OpenAI is moving from a curated, high-spending advertiser base toward a broader, performance-oriented market that mirrors Google and Meta's advertiser ecosystems.
Will ChatGPT ads hurt user trust in the AI?
This is the most significant long-term risk in OpenAI's advertising strategy. ChatGPT's value to users derives from its perceived credibility as a neutral information source. Advertising introduces a commercial incentive that, if poorly managed, could erode that credibility. OpenAI's current mitigation strategy relies on clear 'Sponsored' labeling and category separation between paid placements and organic AI responses. The company reports no impact on privacy-related trust metrics since launch. However, the trust risk is structural and compounding: as advertising volume scales and financial pressure to maximize revenue per query increases, the temptation to blend paid and organic placements — as search engines have gradually done over two decades — becomes a meaningful design risk. The brands most at risk are those whose sponsored placements appear in queries where the user's intent and the advertiser's product don't closely match. Mismatched placements are the fastest way to make users distrust the AI's organic recommendations, which would undermine both the advertising product and ChatGPT's core value proposition simultaneously.
Related Articles
Topics: Distribution & Strategy, AI & Machine Learning, Growth Marketing, Pricing Strategy
Browse all articles | About Signal