The app suggestions that sparked outrage in December were just the warm-up. Six weeks ago, OpenAI insisted the Target shopping prompts appearing in ChatGPT weren’t ads. They were “app suggestions,” the company said, part of an SDK partnership, not a monetization play. Users weren’t buying it. Now, neither is OpenAI. On January 16, 2026, the company officially announced it would begin testing advertisements in ChatGPT for free and Go-tier users in the United States. The same company that called ads “unsettling” and a “last resort” just three years ago is now building an ad platform from scratch.
The December Controversy That Foreshadowed Everything
The warning signs appeared in early December 2025. Benjamin De Kraker, a former xAI employee and paying ChatGPT Plus subscriber, posted a screenshot that went viral: he was asking about Windows BitLocker security, and ChatGPT responded with a prompt to “Shop for home and groceries. Connect Target.” The backlash was immediate. De Kraker wasn’t subtle: “I’m in ChatGPT (paid Plus subscription), asking about Windows BitLocker, and it’s showing me ADS TO SHOP AT TARGET. Yeah, screw this. Lose all your users.”
OpenAI’s response was a masterclass in semantic gymnastics. Daniel McAuley, OpenAI’s data lead for ChatGPT, jumped in to clarify: “This is not an ad (there’s no financial component). It’s only a suggestion to install Peloton’s app.” De Kraker wasn’t convinced: “When brands inject themselves into an unrelated chat and encourage the user to go shopping at their store, that’s an ad. The more you pretend this isn’t an ad because you guys gave it a different name, the less users like or trust you.”
Even Mark Chen, OpenAI’s Chief Research Officer, admitted they “fell short,” acknowledging that “anything that feels like an ad needs to be handled with care.” Nick Turley, head of ChatGPT, tried damage control: “There are no live tests for ads—any screenshots you’ve seen are either not real or not ads. If we do pursue ads, we’ll take a thoughtful approach.” Six weeks later, that “if” became a “when.”
The Code Red That Changed Everything
Between the December controversy and January’s ad announcement, something significant happened inside OpenAI: Sam Altman declared a “code red.” On December 1, 2025, Altman sent an internal memo warning employees that the company was “at a critical time for ChatGPT.” Google’s Gemini 3 had just launched to widespread praise, outperforming ChatGPT on key benchmarks. Salesforce CEO Marc Benioff publicly announced he was switching from ChatGPT to Gemini. Anthropic’s Claude was gaining ground with enterprise customers.
The code red meant deprioritizing everything except core ChatGPT improvements, including advertising. Teams working on ads, shopping agents, health assistants, and a personalized feature called “Pulse” were told to stand down. OpenAI released GPT-5.2 (internally codenamed “Garlic”) in mid-December as its counterattack, reclaiming the top spot on several benchmarks. The eight-week code red was scheduled to end in late January. And right on schedule, with ChatGPT’s position somewhat stabilized, OpenAI pulled the trigger on ads.
The Financial Pressure Behind the Pivot
The numbers explain the urgency. OpenAI is burning cash at a staggering rate:
- •$143 billion in projected negative cumulative free cash flow between 2024 and 2029
- •$8 billion lost in the first half of 2025 alone
- •$1.4 trillion committed to AI infrastructure over the next eight years
- •$200 billion in revenue needed by 2030 just to turn a profit
And here’s the brutal math: of ChatGPT’s 800 million weekly users, only about 5%—roughly 35-40 million—actually pay for a subscription. Sam Altman himself admitted that the $200/month Pro plan loses money because users consume more compute than anticipated. Deutsche Bank analysts put it bluntly: “No startup in history has operated with losses on anything approaching this scale. We are firmly in uncharted territory.” Meanwhile, Google generates over $74 billion in quarterly ad revenue. Meta pulls in more than $50 billion. The math is obvious: if you can’t convert free users to paid, you monetize them with ads.
What ChatGPT Ads Will Actually Look Like
According to OpenAI’s announcement, here’s what’s coming:
Where ads appear: At the bottom of ChatGPT’s answers, clearly labeled and separated from the organic response.
Who sees them: Free-tier and ChatGPT Go ($8/month) users in the U.S. logged-in adults only.
Who doesn’t: Plus ($20/month), Pro ($200/month), Business, and Enterprise subscribers remain ad-free.
What’s excluded: Users under 18, plus conversations about politics, health, and mental health.
OpenAI laid out five principles governing their approach:
- Mission alignment: Ads support making AI accessible
- Answer independence: Ads won’t influence ChatGPT’s responses
- Conversation privacy: User data isn’t sold to advertisers
- Choice and control: Users can disable personalization and clear ad data
- Long-term value: No optimization for time spent in app
Fidji Simo, OpenAI’s CEO of Applications (and notably, a former Facebook and Instacart executive), emphasized on X: “Ads will not influence the answers ChatGPT gives you.”
The Trust Problem OpenAI Can’t Escape
Privacy advocates aren’t reassured. Miranda Bogen, director of the Center for Democracy and Technology’s AI Governance Lab, argues that even without direct data sharing, targeted advertising creates dangerous incentives. “Even if AI platforms don’t share data directly with advertisers, business models based on targeted advertising put really dangerous incentives in place when it comes to user privacy,” Bogen said. “This decision raises real questions about how business models will shape AI in the long run.”
The concern is straightforward: people use ChatGPT for deeply personal tasks—from therapy-adjacent conversations to career advice to medical questions. When that trust relationship becomes a monetization opportunity, the dynamics change. One Slashdot commenter put it simply: “Web search is trending on a path similar to TV. At first, TV was free with no ads. Very quickly ads were introduced. Then came cable TV where you could pay and watch ad-free. Fairly soon after that cable TV had ads too.”
The Competitive Context
OpenAI isn’t making this move in a vacuum. The AI landscape has shifted dramatically:
- •Google’s Gemini grew from 450 million to 650 million monthly active users since July 2025
- •Anthropic now holds 32% of enterprise market share versus OpenAI’s 25%—a complete reversal from one year prior when OpenAI had 50%
- •Prediction markets now give Google 87-92% odds of having the leading model by year’s end
Three years ago, Google declared its own “code red” in response to ChatGPT’s launch. Larry Page and Sergey Brin reportedly came back to review code personally. Now the positions have completely flipped.
What Happens Next
The ad rollout begins “in the coming weeks” for U.S. users. OpenAI says it will learn from feedback and iterate, the same language every platform uses before ads become increasingly intrusive. The company is betting that clear labeling, topic exclusions, and privacy controls will differentiate ChatGPT ads from the surveillance capitalism model that defined the last era of digital advertising.
But the structural pressures are obvious. OpenAI needs $200 billion in revenue by 2030. Subscriptions alone won’t get there. If ad revenue becomes material to the business, the incentives to expand that revenue—more placements, more targeting, more engagement optimization—become harder to resist. As one analyst noted: “The question isn’t just whether ads influence today’s responses. It’s whether they’ll subtly shape recommendations months or years from now. Will ChatGPT still suggest the best product, or the one that pays the most?”
For now, paying subscribers can avoid ads entirely. But the precedent is set. The company that promised to put users first is now building an ad platform on top of the most personal digital assistant most people have ever used. The December “app suggestions” weren’t ads. Until they were.