AI is reshaping the open internet. Is the buy side paying attention?

A U.K. government report, predicted to clarify how AI companies can use publishers’ and artists’ content, landed this week.

Officials scrapped plans that would have allowed AI companies broad rights to train on copyrighted content unless a creator specifically opted-out, after overwhelming opposition from the creative industries. But instead of proposing an alternative, they called for more evidence and further monitoring.

Publishers blocked the outcome they feared most. They just didn’t get what they wanted either. Still, many welcomed the news.

But while coverage of AI chatbots has largely focused on the impact on publishers and creatives, the effects are increasingly landing on advertising’s buy-side too.

AI chatbots aren’t just reducing traffic — they’re reshaping how value shows up on the open internet, making premium ad inventory both more visible and more scarce. As chatbots siphon away low-intent, commoditized traffic, what remains are high-attention environments like trusted news, where engagement — and advertising impact — are harder to replicate.

The catch is that as premium environments become easier to identify, fewer quality placements outside the walled gardens could mean tighter supply and a corresponding impact on media buyers’ negotiation power.

“Concentration creates dependencies, which usually leads to higher prices,” said Michael Beuth, managing director at Mediaplus Group.

A bit of context

Last month, the BBC, Financial Times, The Guardian, Sky News and The Telegraph formed Standards for Publisher Usage Rights (SPUR), a coalition to create global standards that ensure AI companies pay for the journalism they train on. The move would have been unthinkable even a year ago. Now it looks necessary.

Two of SPUR’s founders — The Guardian and Financial Times — already have individual licensing deals with OpenAI. The fact that they joined forces shows the scale of the issue, but it also shows that these publishers know the true value of their content and will fight for it.

The move comes as AI scraping activity on news sites jumped 55% in Q4 2025 compared to the previous quarter, according to a report by TollBit. Click-through rates to publisher sites dropped significantly in 2025, regardless of whether publishers had licensing deals with AI companies.

How media buyers see it

Chris Milano, global vice president of supply at Assembly Global, mapped out the risk. “The risk of a ‘zero-click internet’ is people will stop interacting with publishers directly, and as a result, less advertising dollars go to those publishers,” he said. “Less visits equates to less ad opportunities.”

Milano hopes new commercial arrangements can slow this slide. “For the sake of everyone, I hope we can see new models that reward publishers for content licensing and potentially even set up revenue share models where payments can be split between LLM providers and publishers,” he said.

Beuth, on the other hand, sees the media mix reshuffling. “Completely new partners are currently emerging — for example, AI systems like ChatGPT increasingly crawl and use content from platforms like Reddit or Roblox, which opens up entirely new advertising opportunities,” he said. Popular platforms like Meta and YouTube “may face new forms of competition from previously unexpected sources driven by AI.”

As advertising fragmentation gains momentum, Beuth said that buyers should beware the inertia of concentrating spend. “[...] Simply increasing platform budgets is reaching its marginal utility — more platform investment does not necessarily bring more benefits,” Beuth said.

The other regulatory track to watch

The copyright report is not the only regulatory intervention worth watching.

The U.K.’s Competition and Markets Authority (CMA) is separately reviewing Google’s AI Overviews under new digital markets powers. It has proposed requirements that would let publishers opt out of Google's AI-generated search summaries without losing visibility in organic results.

Both tracks point in the same direction: making it harder for AI companies to use publisher content without compensation. That may not reverse traffic declines, but it could protect the business models of the publishers behind some of the open web’s most premium and effective ad inventories.

Regardless of how the measures play out, one point is clear: Now more than ever, as AI reshapes content on the web, advertisers should pay attention to diversification in their media plans.

“It’s … about having the right channel for each part of the funnel,” Milano said. “Diversified plans reduce frequency waste, reach incremental audiences and create more resilient performance.”