Google Explores Opt-Out Options for AI Overviews Following UK Regulatory Pressure
The UK’s Competition and Markets Authority (CMA) has issued a series of proposals that could fundamentally break the "scrape-or-sink" cycle currently dominating the web. The core requirement is simple: Google must allow publishers to opt out of having their content digested for "AI Overviews" without being buried in traditional search rankings.
Google says it is "exploring updates" to grant this control. For decades, the internet functioned on a clear quid pro quo: Google crawls a site, and in exchange, the site gets a click. AI Overviews have shattered that pact. By serving instant, AI-generated answers at the top of the page, Google keeps users within its own ecosystem, effectively cannibalizing the very sources it relies on for information.
The DMCC Hammer: A Push for a Fairer Deal
These proposals aren't just polite suggestions. They are backed by the UK’s new Digital Markets, Competition and Consumers Act (DMCC), which grants the CMA the legal teeth to designate Google as having "Strategic Market Status." This allows regulators to bypass years of litigation and impose direct conduct requirements on Mountain View.
Recent industry data paints a grim picture. Since the rollout of AI Overviews, some SEO agencies have reported click-through rate collapses of 20% to 60% for informational queries. Sarah Cardell, the CMA’s chief executive, described the new measures as "targeted and proportionate." The goal is to stop Google from forcing a "take it or leave it" ultimatum on a media landscape that is already on life support. Will Hayter, executive director for digital markets, argues that decoupling AI scraping from search visibility is the only way to ensure the web remains a viable place for human-led journalism.
Technical Barriers and Fragmented Control
Google’s response highlights a defensive posture. Ron Eden, Google’s principal for product management, warned that new controls must not "break search" or create a "confusing experience." But for publishers, search is already broken.
nosnippet tag blocks AI Overviews, but it also strips the site of its traditional search snippet. Visibility plummets. Google-Extended stops Gemini from training on a site's archives, but it doesn't touch the real-time scraping used for search results. The only other option is the "nuclear" route: blocking the Googlebot crawler via robots.txt. This removes the site from search entirely. It isn't a choice; it’s a suicide pact.The "So What": An Impossible Choice for Publishers
For digital publishers, the CMA’s intervention addresses an impossible choice: allow Google to strip-mine your reporting for a zero-click AI summary, or opt out and watch your brand authority vanish. This isn't a strategic dilemma; it’s a fight for survival. If a premium news outlet opts out under the current system, Google’s AI doesn't stop providing answers—it simply sources them from lower-quality, less-reliable competitors who are willing to trade their IP for scraps of visibility.
E-E-A-T (Experience, Expertise, Authoritativeness, and Trustworthiness) was supposed to be the publisher's shield. Instead, Google has used it as a filter to find the best content to summarize for free. Publishers are now forced to weigh the prestige of being a "trusted source" against the hard reality of evaporating ad revenue and shrinking newsrooms.
The consultation on these CMA proposals closes on February 25, 2026. While Google claims to be exploring "granular controls," the power dynamic is shifting. The era of Mountain View setting the terms of the open web unilaterally is hitting a wall in London. Whether Google can evolve its business model without destroying its own supply chain remains the defining question of the year.