UK regulators push Google to separate AI scraping from search rankings

Google has been told to split AI scraping from search ranking, and that is the real pressure point in Britain’s latest move on AI search. The Competition and Markets Authority required the company to let publishers stop their content from being used for AI summaries and model training without disappearing from regular search results, while also pushing Google to add clearer links when publisher material appears in AI-generated answers.

The conduct requirement landed on June 3, 2026, under the UK’s digital markets regime, after the CMA had already designated Google as having strategic market status in general search and search advertising services. British officials said the new rules give publishers, including news organizations, “effective tools” to stop their work from powering AI search features and strengthen their hand in negotiations with Google. That matters because the CMA’s own decision materials flagged a familiar problem: AI Overviews sit so prominently on results pages that opt-outs can be hard to exercise in practice.

Google responded by adding a Search Console control that lets website owners choose whether their pages can appear in, and help ground, Google’s generative AI Search features, including AI Overviews, AI Mode and AI Overviews in Discover. Sites that opt out will still appear in regular Google Search results, but Google said they will not receive traffic or impressions from those generative AI features. The setting does not apply to the Gemini app, and reports around the rollout said the opt-out becomes enforceable on June 17, 2026.

For publishers and brands, the economic logic changes fast when permission and ranking are no longer bundled together. A publisher can now think separately about distribution, attribution and reuse, deciding whether to let high-value reporting, reviews or product pages feed AI answers, or keep that material closer to the site where engagement and conversions happen. Clearer attribution also makes it easier to measure when AI search is driving visits, leads or sales, instead of leaving AI referrals buried inside a black box.

The policy shift is landing at the same time vendors are racing to make AI systems more operational. AdRoll launched an MCP Server built on the open Model Context Protocol standard, connecting campaign data with tools such as Claude, ChatGPT, n8n, Microsoft Agent/Copilot Studio and Cursor, and it is meant to support both insight and action, not just reporting. AdLift also folded Claude into Tesseract to analyze web traffic patterns and identify where software tools crawl and gather data, while Afiniti released a centralized decision platform for business communication centers. The direction is clear: regulators are trying to give publishers leverage over AI use, while martech platforms are building the machinery to track and act on AI visibility.