Neil Young

In a world first, the UK’s Competition and Markets Authority (CMA) has ordered Google to let publishers block their content from AI-generated search summaries. 

Until now, there was no way to keep content out of AI features without losing search visibility altogether. The ruling hands publishers much-needed leverage to push for paid licensing deals, and keep premium content viable in a zero-click world.

But the controls won’t be limited to publishers. When Google rolls them out, any website owner will be able to manage how their content appears in its AI features, which means brands, too, can choose to disappear from Google’s AI results.

Tempting as that may be, my advice is simple: don’t. 

Here’s why.

Google’s AI Mode is now the destination

On 10 July, Google made its AI answer the default for every search worldwide, pushing the traditional list of blue links below the fold for the first time in the search engine’s history. Google’s AI Mode has passed one billion monthly users, and more than 60% of searches now end without a click. 

The AI answer is becoming the destination and search is no longer the first step to a brand’s website. Opting out of Google’s AI summaries won’t redirect people to a brand’s own content – it will simply gift the field to competitors. As AI search keeps growing, cutting brand visibility is an own-goal.

There’s an upside to staying in, too. Google’s AI answers must now carry clear links to their sources, so brands that stay opted in can win traffic from an audience that arrives already informed and engaged.

Opting out would block access to a brand’s owned content, but it wouldn’t necessarily remove the brand from AI-generated answers. Ask AI about a brand’s market or category and the answer won’t be empty. It will be assembled from all other available sources: competitors, critics, and whatever third-party content ranks highest. Imagine a rival has published a listicle ranking the brand a middling third or fourth in the category. Because AI models love a listicle – simple structure, ready-made rankings – that verdict is exactly the kind of thing they’ll lift and repeat as fact. Now, picture that answer in front of a potential customer, investor, or journalist.

What to do instead

The question isn’t whether to opt out – it’s whether you know how AI already sees you. Does Google’s AI know who you are? And when it describes you, does it get you right? Many brands can’t say. And it isn’t only Google – ChatGPT, Claude and others are also fielding questions about your brand. 

This is why at flightfirst group we built Signal, our AI visibility tracker. It maps how AI describes a brand across the major answer engines, what’s shaping those answers, and what to do about it. The things AI draws on are the things that have always made or broken a reputation: what trusted outlets say, and whether a company tells a clear, consistent story about itself. What’s new is the technical layer. AI engines favour content they can read cleanly, so how a site is built matters as much as what it says. Pages that answer real questions in plain language get cited – the rest may get skipped, and the engine is likely to infer or fill gaps from elsewhere.

Recent research by Muck Rack shows that 84% of the links AI cites come from earned sources such as news articles, third-party analysis and other non-paid sources. The discipline that has always shaped how the media understands an organisation now shapes how AI understands it too. But earned coverage alone can’t carry the load, and if publishers start withholding content from AI, the coverage it can draw on narrows further. 

Staying visible takes a spread of channels working together: earned, owned and third-party platforms from LinkedIn to Wikipedia.

The UK is now the test bed for AI visibility on Google. Other territories and platforms will likely follow. The advantage will flow to brands that audit what AI says about them, and act on it – before the story gets written without them.