What the CMA Actually Did

Britain's Competition and Markets Authority issued legally binding conduct requirements on Google on June 3, 2026, under the UK's digital markets competition regime. According to the CMA, this is a "world first" — the first binding ruling of its kind imposed on a major tech platform in the UK.

The rules require Google to give publishers a real opt-out: block your content from AI Overviews, AI Mode, and AI Overviews in Discover. Block it from being used to fine-tune Google's AI models. And force Google to actually link back to publisher content in AI-generated results — not just use it silently.

CMA Chief Executive Sarah Cardell put it plainly: "With features like AI Overviews rapidly reshaping online search, it is crucial that content publishers, including news organisations, have appropriate bargaining power over how their content is used."

How We Got Here

This didn't happen overnight. The CMA designated Google as having "strategic market status" in UK search back in October 2025 — a formal finding of substantial and entrenched market power that unlocked the authority to impose these targeted rules. Draft proposals were floated in January 2026. Today's ruling is the follow-through.

Timing matters here too. The CMA said it was also responding to Google's May 2026 announcement that it planned to dramatically expand AI's role in its search platform. According to Silicon Republic, the regulator explicitly stated today's requirements will apply to those upcoming changes. Google doesn't get to outrun the rules by announcing new features.

The Mechanism: A Toggle in Search Console

Google's response wasn't defiance — it was compliance, at least in form. Google's general manager of search ecosystem, Mrinalini Loew, announced in a blog post that Google is rolling out a new toggle in Search Console allowing website owners to manage how their content appears in generative AI search features.

Sites that opt out won't receive traffic or impressions from AI features, according to Google. Critically, Google says the opt-out setting will NOT affect rankings in standard search results outside of AI features. That's important — it means publishers aren't being punished in organic search for refusing to feed the AI machine.

Google is testing this with a "subset of website owners in the UK" first, per The Verge, with plans to roll it out globally after testing.

New Search Console performance insights are also coming — showing publishers which of their pages are appearing in AI-generated responses, and in which countries.

The Actual Problem This Is Solving

The CMA previously found that news publishers suffered a measurable drop in traffic after Google rolled out AI Overviews. According to AP reporting via WRAL, fewer users click through to original articles when a summary is sitting right there at the top of the page.

Google takes your content, summarizes it, puts the summary at the top of search results, and your traffic drops. Then Google profits from AI features powered by your work. Until today, publishers had zero leverage to say no.

AI Overviews now has more than 2.5 billion monthly users, according to figures cited by Asharq Al Awsat. That's the scale of the content pipeline publishers have been involuntarily feeding.

What Mainstream Coverage Is Getting Wrong

Most coverage is framing this as a Google-vs.-regulators story. It's not. It's a property rights story.

Publishers — from individual bloggers to major newsrooms — created content that has real value. Google built a product that extracts that value without compensation or permission. The opt-out requirement doesn't fix the underlying theft that already happened. It just stops the bleeding going forward.

The Verge framed this primarily as a win for publishers and a transparency measure. Fair enough. The harder question remains unasked: should opt-out even be the standard here, rather than opt-in? Requiring publishers to actively choose to protect their work places the burden on the victim, not the extractor.

An opt-in model — where Google must get affirmative permission before using content for AI — would be a fundamentally stronger protection. Regulators stopped short of that.

What This Means for Regular People

If you run a website, a news outlet, a blog, or any online publication with content indexed by Google in the UK — you now have a real legal mechanism to keep your work out of AI Overviews and Google's AI training pipeline. That's a genuine shift in bargaining power.

If you're an American publisher watching this from across the Atlantic: the EU is investigating similar issues, and there's zero equivalent protection in the US right now. Congress hasn't moved on this. The FTC hasn't moved on this. Britain just lapped the United States on protecting content creators from Big Tech extraction.

For consumers, attribution requirements mean AI-generated results should actually link back to sources — theoretically making it easier to verify where information comes from. Whether Google implements this meaningfully or buries the links in fine print will determine its actual value.

Google is a $2 trillion company. A Search Console toggle isn't a revolution. It's a start — and regulators forcing it is the only reason it exists at all.