OpenAI's advertising experiment inside ChatGPT crossed the $100 million annualized revenue threshold in just six weeks, according to a company spokesperson quoted by Reuters on March 26. The milestone, disclosed as the company simultaneously announced international expansion into Canada, Australia, and New Zealand, signals that demand from advertisers for access to conversational AI audiences has outpaced even aggressive internal projections. It also raises fresh questions about whether the platform can sustain that trajectory without compromising the user trust that underpins its appeal.

The numbers are difficult to ignore. Roughly 85% of ChatGPT users are currently eligible to see advertisements, yet fewer than 20% encounter them on any given day, according to the same spokesperson. That gap between eligibility and exposure suggests OpenAI has deliberately restrained ad load during the pilot phase - a strategy that leaves considerable room for monetization growth within the existing user base before any international rollout begins. More than 600 advertisers have now joined the program, and nearly 80% of small- and medium-sized businesses have expressed interest in buying ChatGPT placements, according to Reuters.

The advertising test launched on February 9 for logged-in adult users in the United States on the Free and Go subscription tiers. Plus, Pro, Business, Enterprise, and Education subscribers remain exempt. According to OpenAI's own update published March 26 on its blog, "We're seeing no impact on consumer trust metrics, low dismissal rates of ads, and ongoing improvements in the relevance of ads as we learn from feedback." The company characterized these signals as justification for moving into the next phase of its pilot.

From code red to $100 million

The speed of this commercial traction stands in stark contrast to the uncertainty that surrounded OpenAI's advertising ambitions just months ago. On December 2, 2025, CEO Sam Altman issued an internal "code red" directive that paused advertising development entirely. The company redirected all resources toward improving ChatGPT's core functionality as Google's Gemini AI gained competitive ground, surpassing OpenAI on industry benchmark tests. Advertising, along with AI agents for health and shopping and a personal assistant called Pulse, was explicitly shelved.

That freeze lasted roughly six weeks. On January 16, 2026, OpenAI reversed course and formally confirmed its plans to test advertising within ChatGPT, ending months of speculation that had built up throughout 2025. The company had posted a job listing in September 2025 seeking a Growth Paid Marketing Platform Engineer, appointed Omnicom Media Group's PHD as its global media agency of record, and held internal discussions about sponsored content formats - all while publicly maintaining that advertising was a lower priority than product development.

By February 6, OpenAI had begun accepting advertisers. Confirmed early participants included Target, Ford, Mrs. Meyer's, and Adobe, with major holding companies WPP Media, Omnicom, and Dentsu among launch partners. The minimum commitment was set at $200,000, and the CPM rate landed at approximately $60 - a figure comparable to NFL broadcast inventory and premium streaming placements, and more than three times the prevailing CPMs on Meta's platforms.

The first independently verified ad sightings did not arrive until around February 20, when search intelligence firm Adthena analyzed more than 500 prompts and found advertisements appearing in roughly 0.8% of responses. Confirmed advertisers at that point included Expedia, Qualcomm, Best Buy, Enterprise Mobility, and The Knot Worldwide. That 0.8% frequency was not a bug. It reflected the deliberate restraint OpenAI maintained during the initial testing window.

How the ad system works

According to OpenAI's official blog post - originally published February 9, 2026, and updated March 26 - advertisements appear at the bottom of ChatGPT responses when a relevant sponsored product or service matches the conversation topic. The system determines which ad to display by matching advertiser submissions with the subject of the user's current conversation, past chat history, and previous interactions with ads. If a user is researching recipes, for example, meal kit or grocery delivery advertisements may appear. When multiple advertisers compete for the same placement, the system selects the one deemed most relevant to the ongoing chat.

Ads are clearly labeled as sponsored and visually separated from organic answers. According to the same blog post, "Ads do not influence the answers ChatGPT gives you. Answers are optimized based on what's most helpful to you." The company pledges that advertiser conversations remain private - marketers receive only aggregate performance data such as views and clicks, with no access to individual chats, chat history, memories, or personal details.

OpenAI has also established exclusion zones. Advertisements will not appear in accounts where users indicate they are under 18 or where the system predicts underage usage. Ads are also excluded from appearing near sensitive or regulated topics, including health, mental health, and politics. Users retain the ability to dismiss ads, share feedback, learn why a particular ad was shown, delete ad data with a single action, and manage personalization preferences at any time. Those who prefer no ads at all can upgrade to Plus or Pro subscriptions, or opt out of ads on the Free tier in exchange for fewer daily free messages.

These technical safeguards matter because the advertising model introduces structural tensions that did not exist under a purely subscription-based approach. Every advertising-supported platform eventually faces pressure to increase ad load, expand targeting capabilities, and maximize the time users spend on the product. OpenAI has explicitly stated that it "does not optimize for time spent in ChatGPT" and will "prioritize user trust and user experience over revenue." Whether that commitment survives contact with quarterly revenue targets remains to be seen.

International expansion and self-serve tools

The March 26 announcement confirmed that OpenAI will begin expanding the advertising pilot beyond the United States in the coming weeks, starting with Canada, Australia, and New Zealand. According to the updated blog post, the company intends to "roll this out thoughtfully in each market, learn from real-world usage, and adjust as we go." The stated hope is to expand to many more markets before the end of 2026.

Asad Awan, who leads monetization at OpenAI, shared additional context through a LinkedIn post published shortly after the announcement. He described early signals as "strong" and noted that the company was learning from usage patterns and improving relevance over time. Awan has been a central figure in shaping OpenAI's advertising strategy since outlining a longer-term vision in a February 9 podcast where he suggested that small businesses could eventually create and manage campaigns by simply prompting ChatGPT - a vision that drew immediate skepticism from agency professionals.

According to Reuters, OpenAI plans to launch self-serve advertiser capabilities in April. This represents a significant infrastructure milestone. The company had already begun testing an Ads Manager with a small group of partners earlier in March, providing a dashboard for marketers to run, monitor, and optimize campaigns in real time. Early testers received weekly CSV reports containing performance data on clicks and impressions. The transition to self-serve access would dramatically lower the barrier to entry for smaller advertisers who cannot meet the current $200,000 minimum commitment, potentially unlocking the SMB demand that the company's own data suggests is substantial.

Reuters also reported that David Dugan, a former Meta ads executive, was named to lead OpenAI's global advertising solutions team earlier in the week of March 26. The hire signals that OpenAI is building out its advertising leadership with experience drawn from the platforms it seeks to compete against.

The attribution gap

For all the commercial momentum, significant structural challenges remain. ChatGPT still lacks a pixel or any standard attribution system that would allow advertisers to measure conversions with the precision they expect from established platforms like Google or Meta. This gap between what advertisers can prove from a ChatGPT campaign versus what they can verify on other platforms is substantial.

Criteo's entry as the first formal ad tech partner on March 2 introduced some of that infrastructure. The Paris-headquartered company, which activates more than $4 billion in annual media spend across approximately 17,000 advertisers, connected brands to ChatGPT's Free and Go tiers in the United States. Internal data from a sample of 500 Criteo retailers observed in February 2026 found that users referred from LLM platforms like ChatGPT convert at approximately 1.5 times the rate of other referral channels. The sample is narrow - 500 retailers from a single month - but the directional signal is what the industry is watching.

However, not all early conversion data has been encouraging. Walmart disclosed in March 2026 that conversion rates for products sold directly inside ChatGPT were three times lower than those requiring users to click through to Walmart's own website. That contradiction - Criteo seeing higher conversion rates while Walmart sees lower ones - suggests the advertising experience varies significantly depending on product category, purchase complexity, and the specific path from chat to transaction.

The $60 CPM rate also puts pressure on advertisers to demonstrate return on investment from a channel that does not yet offer the granular measurement tools available elsewhere. Premium streaming inventory and NFL broadcasts - the closest pricing comparisons - come with established audience measurement frameworks and decades of advertiser experience. ChatGPT offers neither.

Competitive context and industry implications

OpenAI's advertising acceleration occurs within a competitive landscape that is shifting rapidly. Google has systematically expanded advertising across its own AI-powered search features, rolling out ads in AI Overviews to 11 countries beyond the United States in December 2025. The company has also introduced shopping ad formats within AI Mode and continued testing sponsored placements across various AI surfaces. These developments mean that advertisers evaluating ChatGPT placements are not comparing them against static alternatives - the entire category of AI-powered advertising is expanding simultaneously.

Anthropic took a different approach entirely. The company committed in February 2026 that Claude would remain advertisement-free across all subscription tiers, launching a multimillion-dollar Super Bowl advertising campaign that explicitly criticized OpenAI's decision to introduce ads. Sam Altman responded by characterizing Anthropic's Super Bowl ads as "dishonest", arguing that OpenAI's advertising-supported free tier provides access to users who cannot afford premium subscriptions. The exchange crystallized a fundamental strategic divergence in the AI industry: subscription-only models that charge for access versus advertising-supported models that monetize attention.

For the marketing community, the implications extend beyond a single platform. The emergence of conversational AI as an advertising channel introduces measurement challenges that existing frameworks were not designed to handle. Traditional digital advertising relies on impression-level reporting, click tracking, pixel-based attribution, and deterministic audience matching. ChatGPT offers contextual relevance derived from conversation content, but the measurement infrastructure that would give performance advertisers confidence in the channel is still being built.

OpenAI's internal documents, leaked earlier in 2026, projected $1 billion in advertising revenue for 2026 and $25 billion by 2029. For context, Amazon - now the third-largest digital advertising company globally - generated $56 billion in ad revenue last year with assets including NBA rights, Prime Video, and a vast commerce data reservoir. Whether OpenAI can approach those projections depends on solving the attribution problem, expanding beyond the current pilot's deliberately restrained ad load, and maintaining the user trust metrics that the company cites as evidence the pilot is working.

The controlled rollout also reflects lessons from social media's early advertising experiments. Former OpenAI researcher Zoë Hitzig resigned on February 11, publishing a New York Times commentary titled "OpenAI Is Making the Mistakes Facebook Made. I Quit." She warned that economic pressures from an upcoming IPO - planned for the fourth quarter of 2026 - could create strong incentives to override the company's own privacy commitments. OpenAI's response has been to maintain tight control over partner messaging. Advertisers participating in the pilot cannot discuss pricing, performance data, or internal results publicly and must route sensitive questions back to OpenAI, with all external materials requiring pre-approval.

What happens next

The April launch of self-serve advertiser capabilities will be the next major test. Lowering the entry barrier from $200,000 minimum commitments to accessible self-service tools could substantially increase the advertiser count beyond the current 600. The international expansion to Canada, Australia, and New Zealand introduces new regulatory environments and consumer expectations that may require adjustments to the ad format, targeting methodology, and privacy protections.

OpenAI's financial trajectory makes advertising increasingly important. The company must grow revenue to roughly $200 billion to turn a profit by 2030, according to projections reported in December 2025. Subscription revenue alone - even with ChatGPT's 700 million weekly active users - is unlikely to reach that scale. Advertising provides a complementary revenue stream, but only if the platform can demonstrate measurable value to advertisers while preserving the conversational experience that attracts users in the first place.

The $100 million annualized revenue figure, while commercially significant, represents a fraction of what established advertising platforms generate. Google's advertising revenue exceeded $300 billion in 2025. Meta's advertising business operates at roughly similar scale. Even Perplexity AI, which launched sponsored follow-up questions in November 2024, has been testing premium CPM rates above $50 with its smaller user base. The question is not whether $100 million in six weeks is impressive - it is - but whether the growth can compound as ad load increases, international markets open, and advertisers demand the kind of performance proof that the platform does not yet provide.

OpenAI closed its blog update with a statement that will be tested repeatedly in the months ahead: "What will always remain true: ChatGPT's answers remain independent and unbiased, conversations stay private, and people keep meaningful control over their experience." The advertising industry has heard similar pledges before, from platforms that eventually prioritized revenue growth over user experience. Whether OpenAI breaks that pattern or follows it will shape not just its own trajectory, but the future of how AI platforms are funded.

Timeline

Summary

Who: OpenAI, the artificial intelligence company behind ChatGPT, along with its monetization lead Asad Awan and newly appointed global advertising solutions head David Dugan (formerly of Meta). The pilot involves more than 600 advertisers, including early participants Target, Ford, Mrs. Meyer's, and Adobe, with holding companies WPP Media, Omnicom, and Dentsu as launch partners, and Criteo as the first formal ad tech partner.

What: OpenAI's ChatGPT advertising pilot in the United States surpassed $100 million in annualized revenue within six weeks of launching. The company simultaneously announced international expansion to Canada, Australia, and New Zealand, disclosed that 600 advertisers have joined the program with nearly 80% SMB interest, and confirmed plans to launch self-serve advertiser capabilities in April. Approximately 85% of users are eligible to see ads, but fewer than 20% encounter them daily.

When: The advertising pilot launched on February 9, 2026 for logged-in adult users in the United States. The $100 million revenue milestone and international expansion were disclosed on March 26, 2026. Self-serve advertiser tools are scheduled for April 2026. International pilots in Canada, Australia, and New Zealand will begin in the coming weeks.

Where: The advertising pilot currently operates in the United States, serving ChatGPT's Free and Go subscription tiers. Expansion will extend to Canada, Australia, and New Zealand as the first international markets, with OpenAI indicating plans for additional countries throughout 2026.

Why: OpenAI requires substantial revenue growth to fund the infrastructure costs of operating ChatGPT and developing its AI models, with financial projections requiring roughly $200 billion in revenue by 2030 to achieve profitability. Advertising provides a complementary revenue stream alongside subscriptions, enabling the company to maintain free and low-cost access tiers for hundreds of millions of users. The pilot also addresses advertiser demand for access to conversational AI audiences, with contextual targeting based on chat content offering a differentiated value proposition compared to traditional search and social advertising platforms.

Share this article
The link has been copied!