Google’s Ultimatum: Publishers Must Share Data for AI Overviews or Lose Traffic
Ddos 
As previously mentioned, Google Search is in the process of integrating advertisements into its AI Overviews, which are rapidly evolving into a monetized space—an initiative of critical importance to Google. Consequently, the company has rebuffed efforts by websites and publishers to prevent the use of their data in generating these AI Overviews.
According to internal documents released by Bloomberg, Google had once deliberated offering publishers more granular control, ostensibly allowing them to determine how their website data is used within AI search functionalities. However, the documents suggest this would have been more illusion than autonomy.
These internal files—drafted by senior executives at Google Search and disclosed amid an antitrust lawsuit in the United States—reveal that Google mandated websites either consent to freely sharing their data for AI Overviews or risk forfeiting all Google-sourced traffic.
The documents underscore growing concerns that AI Overviews and other AI-driven features might significantly undermine content-originating sites, as users may find sufficient information in the overview alone, bypassing the need to visit the original websites. This, in turn, would lead to declining traffic and the erosion of monetization opportunities for publishers.
One proposal, marked as a non-negotiable “red line” for Google, would allow publishers to block AI models from accessing their content in real time. However, they would be denied the option to prevent Google from using that same content—once scraped—for training AI Overviews.
Another suggestion, flagged as “potentially unstable,” advocated that Google refrain from implementing any additional control measures—effectively prohibiting publishers from obstructing the AI models’ use of their data. Should publishers disagree, their sole recourse would be to opt out entirely by blocking Google’s crawlers via the robots.txt protocol. Yet, such a decision would cut off all search traffic from Google—a cost few publishers can afford to bear.
A court hearing on May 2nd revealed that publishers are now facing an ultimatum. Although Google introduced an opt-out framework in 2023 for AI training, Google DeepMind’s Vice President clarified that this exemption does not extend to search-centric AI products such as AI Overviews. Therefore, the only way for publishers to avoid inclusion is to fully withdraw from Google’s web crawling.
Further, the documents disclose that when Google launched AI Overviews in 2024, it quietly modified its guidance concerning publisher control—without issuing any public announcement. Internal directives regarding the language of these updates suggest a deliberate attempt to obscure which functionalities publishers had truly opted out of, thereby preventing a clearer understanding of the interplay between Gemini, AI Overviews, and other AI training mechanisms.
In response, Google stated that the document merely reflects early-stage considerations during the company’s development of AI and does not represent final policy decisions. However, Google’s current language is unequivocal: publishers who flag their content as ineligible for AI Overviews and AI training will simultaneously be excluded from all forms of search visibility—an approach that borders on coercion.
Related Posts:
- YouTube Tests AI Overviews for Video Summaries with Gemini
- Your AI Summary Will Have Ads: Google Integrates Sponsored Content into Search
Support Our Threat Intelligence
If you find our CVE report and cybersecurity news helpful, consider supporting our work.
Buy Me a Coffee
PayPal
