Media Briefing: ‘A bitter pill’: ChatGPT ads are coming – where do publishers fit?

‘A bitter pill’

The details on OpenAI bringing ads to ChatGPT are slowly emerging. And publishers are watching closely to see what this will mean for their own businesses.

A few risks are starting to bubble up: more revenue for OpenAI from content ChatGPT pulls from the open web, without a cut going to publishers, and more competition for ad dollars. A few risks to publishers’ businesses are starting to bubble up. Ads coming to ChatGPT likely mean more revenue for OpenAI from content ChatGPT pulls from the open web, without a cut going to publishers. It also means more competition for ad dollars.

“[OpenAI is] cutting into our business model, and they’re monetizing it beyond just subscriptions. So it really does change how that feels for publishers, and it changes the equation of — ‘you’re getting all of this for free from us, and now you’re actually making money directly off it. You’re using our IP to fill out these answers that you’re then getting an advertiser to pay you because they’ll transact there.’ That’s rough. That’s a bitter pill,” said a publishing exec, who exchanged anonymity for candor.

Marcel Semmler, CPO at Bauer Media, said his team is monitoring these changes closely, though they are not at the “panic level.”

What it really signals is that answer engines are becoming monetization platforms, not just utility layers, he stressed. “And once monetization enters, the question shifts from ‘how do answers work?’ to ‘who gets represented, surfaced, and paid?’ The biggest implication is not ads themselves, but distribution power and dependency. If ChatGPT becomes a monetized interface, it will inevitably shape what content is surfaced, how sources are referenced, and how value is captured. That creates both risk and opportunity,” Semmler said.

More revenue for OpenAI without a publisher cut

OpenAI will soon make additional revenue (other than subscriptions) off the back of publishers’ content with ads. But OpenAI has not announced how — or if — publishers get a cut. In response to a request for comment, a spokesperson said OpenAI is focused on testing ads in the U.S. over the next few weeks.

This added monetization layer is another blow to publishers already concerned with OpenAI using their content to power responses in ChatGPT without proper attribution, control or payment.

Another pain point: Publishers’ content fuels ChatGPT without sending traffic back, cutting into ad revenue. 

“It is likely to make the traffic displacement issue even worse,” said the same anonymous publishing exec. “If you’re in a chat and you cannot only ask a question about a product you want to buy, or a trip you’re thinking of taking… and you’re not only getting a thorough answer, but then you’re getting an ad right next to it with a link to buy that product, book that trip, purchase that service? I think it’s just all the more reason why people are not going to ever exit that chat experience and click on the small link that powers part of the answer results.”

This new revenue stream for OpenAI might also mean publishers that have licensing deals with the tech company may want to revisit those agreements to figure out some sort of cut of OpenAI’s ad revenue when they come to the table, especially as those deals come up for renewal soon.

Competition for ad dollars

OpenAI says it has 800 million weekly active users, and ChatGPT ads put it in direct competition for ad dollars with publishers.

However, Scott Messer, principal and founder of publisher consultancy Messer Media believes that won’t dangerously erode open web display budgets for a few years. “Right now, it’s probably experimental budgets. Sure, it [advertisers allocating budgets] may be a million bucks a piece, but it’s not moving the $400 billion ad ecosystem,” he said.

Others are more unnerved about the short term. “There are finite budgets, and that level of engagement is really tempting [for advertisers],” said the same publishing exec who spoke under anonymity.

On the flip side, publishers already have established infrastructures for advertising and long-term relationships with advertisers, as well as robust ad measurement, said Caroline Giegerich, vp of AI at IAB, the media and marketing trade body.

“If you’re a publisher, you want to know, ‘does this have a real, long-term impact to how advertisers are spending on my platform?’” Giegerich said. Publishers will likely be watching their ad businesses closely, keeping an eye on budget allocation, especially around upper funnel awareness budgets, she added.

The anonymous publishing exec was concerned this wouldn’t be enough. “[Publishers’] metrics are really geared towards awareness campaigns, things like click-through rate or viewability. Whereas if you are able to drive transactions from those ads, it becomes much more performance-based, with very granular performance metrics.”

Ads in ChatGPT could “cannibalize intent-based search advertising much more than CPM-based display advertising,” said David Caswell, founder of StoryFlow, a consultancy focused on AI workflows in news production. “The ‘intent’ signal is so much stronger — it’s like having an instant, personal Consumer Reports for everything you could ever buy, no matter how niche.”

Google ads in AI the bigger threat?

Ultimately, the bigger risk may be if Google were to bring more ads into its AI experiences, such as AI Overviews and AI Mode. Google launched personalized shopping ads for retailers in AI Mode in a U.S. pilot program last month.

“Google worries me more [than OpenAI ads] because now they have an incentive to keep the traffic to themselves by enabling ads in AI Overviews,” said Messer. “I don’t think you’re going to see ads go into the Gemini app itself because they don’t need to create friction in that surface area by putting ads in.”

Messer stressed that OpenAI understands how search ads work, but also knows how hard they are to build. By opting for CPM pricing for its ads model, OpenAI avoids early comparisons with Google’s highly transparent CPC-driven search ads, giving it room to test and learn before attempting to build the kind of competitive, auction-based marketplace that search monetization requires, he added. “What I’d love to see is advertisers withhold their dollars from marketing budgets until they know that content is ethically sourced by the LLMs,” said Messer. 

The anonymous publishing exec worried that if Google added more advertising into AI Overviews or AI Mode, it would take up more space both within those experiences and in traditional search results. “I think that will only accelerate crowding out organic results entirely,” they said.

While ads in ChatGPT won’t change publishers’ strategies overnight, they mark “a clear transition,” said Semmler. “These systems are no longer neutral utilities. They’re becoming media platforms. And once that happens, publishers need a seat at the table — not just as content providers, but as partners with clear rules.”

— Jessica Davies contributed to this story.

What we’ve heard

“Last year… companies were still trying to figure out — especially in Washington — how they wanted to show up with a new administration, what issues were going to matter. And so we definitely saw a slower January, February. This year, it really feels like people are ready to go. We’re 70% sold through for our flagship newsletter. We have events that are already starting [in January, when it’s] typically a really slow month of people getting their footing, and that doesn’t really start till mid February. I see this year as a real building-blocks year.”

— Anna Palmer, Punchbowl News CEO 

Numbers to know

$74 million: The estimated amount of funding state governments will provide to local newsrooms in 2026.

18: The number of new contributors CBS News boss Bari Weiss is hiring, part of her plan to shake up the org.

150%: The increase in the number of daily average U.S. users deleting the TikTok app, in the past week.

What we’ve covered

WTF is liquid content?

• Publishers’ adoption of generative AI is reducing the friction between content and format, making it easier for the same story to appear as shorter summaries, audio, or video, often in real time. A text article may soon be more of a vehicle for original reporting, not a final product. 
• This shift has been dubbed “liquid content” — a catch-all term for stories that can be reshaped and tailored to readers, and redistributed across multiple formats with minimal effort. 

Read more here.

How publishers are approaching AI in 2026

• A new Digiday+ Research report surveying 40 publisher professionals explores how publishers are navigating the opportunities and challenges AI brings.
• In Q4 2025, 93% of respondents to Digiday’s survey said that their companies use AI compared to 42% of respondents who said the same in 2022. 50% of respondents said their companies use AI internally.

Read the report here.

Cloudflare’s Human Native acquisition signals a new content economy for publishers

• Cloudflare’s acquisition of AI startup Human Native signals a turning point in how Cloudflare plans to help build an infrastructure for the AI content economy.
• Cloudflare is effectively building an AI licensing stack for its publisher clients. Human Native helps turn publisher content into AI-ready data and makes sure the people who created it get paid.

Read more here.

Forbes tests prediction platform as engagement strategies move past search 

• Forbes has joined the fray of publishers adding prediction markets to their websites as part of a broader strategy to reduce reliance on search traffic and focus more on onsite engagement. 
• The publisher built its own prediction platform – ForbesPredict – which will launch in beta in February. Unlike real-money prediction markets like Kalshi or Polymarket, Forbes’ platform lets readers make forecasts and track results in exchange for tokens. It’s designed for engagement, not wagers. 

Read more here.

Bold Call: AI will rewrite publishers’ websites in 2026

• We’re calling it: 2026 will be the year AI remakes publishers’ websites — and reader experiences. 
• Publishers like Forbes, Newsweek, Time and The Washington Post are already adding AI tools and features to their websites, but this year those AI-driven onsite experiments will only accelerate, with more real-time personalization coming to their sites using AI.

Read more here.

What we’re reading

The Washington Post is reportedly shuttering its sports desk

The Washington Post is planning on making significant cuts to its workforce in the coming weeks, shuttering its sports desk and making major reductions in its foreign department, Puck reported. The cuts could impact as many as 300 employees.

Yahoo launches AI chatbot called Scout

Yahoo has debuted an AI answer engine called Scout on its site and app, allowing users to find information across the web and provide AI summaries, The Verge reported. 

The Detroit News acquired by USA Today Co 

USA Today Co is buying The Detroit News from MediaNews Group, The Detroit News reported. USA Today Co owns another major daily in the metro area, the Detroit Free Press. Terms of the deal were not disclosed.

Google is rewriting publishers’ headlines using AI

Google has updated its framing of using AI to rewrite headlines and snippets in Discover (rather than displaying the ones written by publishers themselves) as a “feature,” rather than just an “experiment,” Nieman Lab reported.

McClatchy newsroom union’s battle over AI 

McClatchy journalists are pushing management to have more guardrails, transparency and human oversight around AI use, citing flawed listicles and threats to their jobs, Columbia Journalism Review reported.