Google's AI Is Scraping Even Sites That Ask to Be Ignored

Google has admitted in federal antitrust court that it effectively bypasses publisher opt-out requests by using web data for its search division’s AI models, even when those publishers have explicitly refused permission for training by Google’s AI lab, DeepMind. This structural loophole allows the company to scrape and utilize content for products like Gemini and AI Overviews, raising significant concerns about consent and the ethical sourcing of data for artificial intelligence development. The implications for open data are profound, as this practice highlights a lack of transparency and user control over how public web content is harvested for commercial AI training. By claiming that the only way to truly opt out is to remove one’s website from Google Search entirely, the company presents an impossible choice that prioritizes data acquisition over publisher autonomy. This undermines the principles of open access, suggesting that visibility and data usage are being leveraged as tools of coercion rather than governed by clear, voluntary standards. This admission strengthens regulatory arguments that Google is abusing its search monopoly to secure free data for its AI initiatives. If Google uses its dominant market position to force publishers into accepting data scraping as the cost of visibility, it may constitute anti-competitive behavior. For the open data community, this serves as a critical warning that without strict regulatory oversight, dominant platforms can erode intellectual property rights and data sovereignty under the guise of technological advancement.

Source: yahoo.com
Published on 2025-05-06

Google's AI Is Scraping Even Sites That Ask to Be Ignored

Related news