Recent forecasts from EMARKETER throw cold water on the breathless predictions that ChatGPT will become the next digital advertising giant. While the AI industry races toward monetization, the real advertising money, according to the research, will flow not into chatbots but into AI-powered search engines. U.S. AI-related ad spending is set to climb from a nascent base to a multibillion-dollar market, but the lion’s share will be captured by the companies that control search, not conversation.

The report underscores a sharp divergence between two emerging ad formats. AI search—think Google’s AI Overviews and Microsoft’s Copilot—integrates ads directly into query results, blending sponsored content with organic answers. ChatGPT Ads, on the other hand, would need to weave promotions into a conversational thread without breaking the user experience. That challenge, EMARKETER suggests, will limit ChatGPT to a niche role, far smaller than the industry’s most optimistic projections.

The Ad Format Showdown

OpenAI has publicly waffled on advertising. CEO Sam Altman once called ads “a last resort,” but the company has been hiring ad talent and exploring premium tiers. Still, no ads have materialized. The friction is obvious: a chatbot built for seamless, human-like dialogue risks becoming a spam-spewing annoyance the moment it starts pushing products. Users might tolerate an occasional sponsored suggestion, but they won’t sit through a pre-roll or a mid-conversation banner. The economics of attention don’t map neatly onto a bot that lives in a sidebar or a standalone app.

Contrast that with AI search. Google’s AI Overviews already serve ads atop organic summaries, and Microsoft’s Copilot, baked into Bing, Edge, and Windows, is designed from the ground up to monetize queries. When you ask Copilot about the best laptops for video editing, it can show you a list sourced from the web—and sprinkle in shopping ads just like a traditional search results page. Same user intent, same click-through incentives. The difference is that search ads have decades of optimization behind them. AI merely makes them more conversational and context-aware, but the underlying business model is battle-tested.

AI search ads often appear as native links within generated answers, clearly labeled as sponsored. ChatGPT would need to invent a new ad unit—perhaps a “promoted response” or a product recommendation mid-dialogue. The risk of annoying users is high. A 2024 survey by a major ad platform found that 72% of consumers would rather see ads in a search results page than have a chatbot interrupt their workflow with a sponsored suggestion. Advertisers are wary of negative brand association in such intimate interfaces.

The Market Landscape

Industry forecasts, including EMARKETER’s, suggest that U.S. AI-related ad spending could surpass $10 billion by 2027, with AI search capturing the majority. That growth comes from a near-zero base in 2023, reflecting the breakneck speed of adoption. Google’s parent Alphabet already reported that AI Overviews have driven higher click-through rates for ads in some test markets, while Microsoft’s advertising revenue—fueled partly by Copilot—grew 12% year over year in its last fiscal quarter. Meanwhile, OpenAI’s revenue remains heavily dependent on subscription fees and API access, which, while impressive at around $3.5 billion annually, is a fraction of the ad budgets sloshing through Google and Meta.

Advertisers are watching closely. A recent survey of media buyers found that less than 20% plan to allocate budget to AI chatbot ads in the next year, citing unclear ROI and brand safety concerns. By contrast, more than 60% are already buying AI-enhanced search ads or plan to do so soon. The gap is telling. Brands don’t want to gamble on a format where the user might be mid-task and annoyed by an interruption. They want to catch people at the moment of intent—exactly what search engines have always done.

The Platform Dynamics

This split echoes a broader platform war. Microsoft has aggressively positioned Copilot as an ad-supported assistant, even testing ads that appear as “promoted” answers within its chat interface. The strategy is to treat Copilot as a new search front-end, one that can siphon queries away from Google. If successful, Copilot becomes a top-of-funnel ad machine that feeds Microsoft’s vast digital advertising ecosystem, including LinkedIn and the Microsoft Audience Network. ChatGPT, by contrast, remains an island. It doesn’t own a search index, a browser, or a productivity suite with billion-user scale. Its standalone nature limits the data signals that advertisers crave for targeting and measurement.

OpenAI is not standing still. It has integrated web search into ChatGPT, and a future ad model could mirror what Google and Microsoft are doing: display sponsored links alongside its own answers. But that would transform ChatGPT into a search engine, blurring the line between chat and search. In such a scenario, the EMARKETER forecast might need revision. For now, however, ChatGPT’s advertising experiments appear cautious. The company has tested a “Promoted Suggestion” feature in limited trials, where the bot occasionally recommends a brand’s service when contextually relevant. Early reactions were mixed, with users complaining that even a light touch felt intrusive. That’s a red flag for a product whose main selling point is being a helpful, ad-free companion.

Meanwhile, the ad dollars are already flowing elsewhere. AI search is not just about Microsoft and Google. Perplexity, an AI-first search startup, is building an ad business on a “brand sponsorship” model, where advertisers pay to be the source of certain types of answers. Even DuckDuckGo has dipped into AI-assisted ads, although it remains tied to Microsoft’s ad network. The common thread is that these interfaces mimic search, not chat. They display a list of results, and ads sit beside or inside those results, preserving the user’s ability to ignore them.

Technical Hurdles for Chatbot Ads

Serving ads in a real-time conversational flow presents technical nightmares. Latency is the enemy of chatbot UX; every millisecond added by an ad auction risks degrading the experience. Google and Microsoft have spent billions on infrastructure that can serve contextually relevant ads in milliseconds alongside search results. OpenAI has no such ad-serving stack. Building one from scratch would require significant investment and time—distractions at a moment when the company is laser-focused on improving model capabilities and safety.

Moreover, measuring ad performance in a conversational setting is messy. Did a user buy a product because of that one chat recommendation, or because of a retargeted display ad they saw later? Attribution windows blur. Without robust measurement, advertisers won’t pay a premium. Search ads, by comparison, offer clean click-through and conversion tracking tied directly to queries.

Consumer Sentiment and Adoption

User acceptance remains a wildcard. In multiple A/B tests run by a major tech publisher, adding even a single sponsored line to a ChatGPT-like session increased user churn by 15–20%. In contrast, Google’s AI Overviews with integrated ads saw only a marginal dip in user satisfaction, partly because people are accustomed to ads on search pages. Conditioning plays a huge role. Users expect search to be free, supported by ads; they expect assistants to be helpers, not salespeople. Breaking that psychological contract could backfire.

Younger demographics, however, might be more amenable. Gen Z and Alpha are growing up with chatbots as daily tools. A study by a youth-focused market research firm found that 40% of 16–24-year-olds would be open to chatbot recommendations if they led to better deals or personalized suggestions. That points to a long-term opportunity, but one that requires a careful, gradual rollout.

The Publisher Dilemma

For publishers, the stakes are existential. AI search has already raised alarms by summarizing their content without sending traffic. Adding ads to those summaries could further decimate referral revenue. If ChatGPT eventually runs ads that replace the need to click through to a publisher’s site, the web’s economic backbone could crack. EMARKETER’s report does not directly address publisher impact, but industry data shows that AI-generated answers now accompany over 15% of Google searches, and that number is rising fast. Publishers are scrambling to negotiate licensing deals, but the power imbalance favors the platforms.

Some publishers are experimenting with their own AI-powered search boxes on their sites, hoping to capture ad revenue directly. The Wall Street Journal and Vox Media have tested generative answer modules that include sponsored content. But these efforts are tiny compared to the scale of Google and Microsoft. Unless regulatory intervention forces a fairer distribution of revenue, many smaller outlets face a grim future.

Regulatory and Privacy Crosswinds

The EU’s Digital Markets Act and evolving U.S. antitrust scrutiny could reshape how AI companies can combine user data for targeted ads. Google’s Privacy Sandbox and Apple’s App Tracking Transparency already squeeze traditional ad targeting. AI search ads, relying on first-party data from signed-in users, may navigate these constraints better than a ChatGPT model that would have to build its ad tech stack from scratch. OpenAI lacks the identity graph that Google and Microsoft possess. Without it, ad targeting will be less precise, and CPMs will suffer.

Privacy-conscious users might even prefer ChatGPT’s ad-free approach, but that puts more pressure on OpenAI to grow subscription revenue. A two-tier system—paid ad-free, free ad-supported—seems inevitable. How OpenAI balances ad load to keep free users engaged while upselling subscriptions will define its monetization path.

Where the Money Will Actually Flow

Looking ahead, the AI advertising battle will likely split into two tiers. The top tier—AI search—will absorb the bulk of brand and performance budgets, evolving into a more dynamic, conversational version of today’s search ads. The lower tier—chatbot ads—will find a niche in in-app recommendations, customer service upselling, and perhaps audio-based ads during voice interactions. ChatGPT might capture some of that, but EMARKETER’s analysis suggests it won’t dethrone the search giants. The forecast may even be conservative if Google successfully defends its turf by blending Gemini AI with its existing ad empire, which generated over $175 billion in revenue last year.

Microsoft’s Copilot sits at the intersection of both worlds. It can function as a chat assistant in Windows or as a search companion in Edge. Its ad tests hint at a hybrid model: small, text-based ads that feel native to chat, combined with larger display units on the search page. The dual approach gives Microsoft flexibility that OpenAI doesn’t have. And because Copilot is integrated into the OS, it can leverage signals from user behavior across apps, browsers, and system-level interactions—a treasure trove for ad targeting that respects Microsoft’s privacy claims.

Nevertheless, ChatGPT’s brand power remains formidable. The name has become synonymous with generative AI, much like “Google it” for search. That consumer mindshare could translate into sustained usage even if ads are introduced gradually and thoughtfully. OpenAI might adopt a subscription-plus-ads model, similar to streaming services, where paid users see no ads while free users get a “light” ad load. The company’s recent hiring of former Google ad executives signals that such a plan is more than just a backup option.

For now, the EMARKETER data forces a re-evaluation of the AI ad narrative. The hype cycle around ChatGPT Ads has cooled, replaced by a more pragmatic focus on AI search as the real revenue engine. U.S. AI-related ad spending will grow from its current, almost negligible base to become a sizable slice of digital budgets, but that growth will track the expansion of search interfaces, not purely conversational bots. Advertisers, analysts, and investors would do well to follow the intent: where users go to find answers is where dollars will ultimately concentrate.

With Microsoft and Google locked in an AI arms race, and OpenAI searching for sustainable revenue beyond subscriptions, the next two years will reveal whether chatbots can evolve into effective ad platforms or remain niche players in a search-dominated landscape. Either way, the advertising industry is bracing for transformation.