UK regulators force Google to give publishers more AI search control

Google is about to give U.K. publishers a new lever inside Search Console: a toggle that lets them opt out of generative AI search features without disappearing from traditional results. The Competition and Markets Authority imposed the conduct requirement on June 3, 2026, and paired it with a demand for clearer attribution when publisher content appears in AI-generated answers. For SEO agencies, that moves AI search from a vague visibility problem to a workflow issue involving rights, traffic control and how clients want their content surfaced.

The policy shift matters because it changes what agencies need to measure. If a publisher can block use in AI Overviews and AI Mode while still ranking in ordinary search, the old assumption that crawl access and rankings move together no longer holds. The CMA, which designated Google as having strategic market status in October 2025, described the move as a world first and said it strengthens publishers’ bargaining position. That means agencies will need to separate organic visibility reporting from AI-answer visibility reporting, especially when clients care as much about attribution and licensing leverage as they do about clicks.

That same split is already showing up in tooling. AdLift says it has integrated the Claude model into Tesseract, its system for analyzing web traffic patterns and identifying where software tools crawl and gather data from websites. AdLift says the setup can read the context, sentiment and intent behind AI-generated brand mentions, while also tracking referral traffic from ChatGPT, Claude and Perplexity. For agencies, that is a signal to broaden crawl intelligence beyond server logs and bot flags. AI systems are now part of the traffic mix, and agencies need dashboards that show not just whether a crawler arrived, but how a model framed a brand and whether that framing led to a visit.

Reporting and client education need to catch up just as fast. Google has said the U.K. opt-out will not be used as a ranking signal in traditional search, which gives agencies a clean talking point for publishers worried about losing blue-link performance while testing AI controls. But the practical conversation is more nuanced: a client may keep its rankings and still decide that AI summaries are diluting attribution or weakening direct traffic. That is where content format starts to matter, because clearer sourcing, cleaner structure and stronger brand identifiers make it easier for publishers to be recognized when AI systems assemble answers.

The same trend is spreading into adjacent platforms. AdRoll has launched a Model Context Protocol server in beta that connects its advertising data repository to external clients including ChatGPT, Claude, Cursor, n8n and Microsoft Copilot Studio. AdRoll says the server can retrieve campaign performance data, draft campaigns and support selected management workflows. Agencies that treat these systems as separate from search will fall behind. The quarter ahead belongs to shops that can read policy changes, crawl behavior and AI-platform integrations as one operating system, not three disconnected trends.