In a remarkable stride towards enhancing online safety, Google announced the blocking of an unprecedented 8.3 billion ads globally in 2025, a significant increase from 5.1 billion in the previous year. This achievement, highlighted in Google's 2025 Ads Safety Report, raises intriguing questions about the effectiveness of its enforcement strategies.
Google credits this surge in blocked ads to its advanced use of AI technology, particularly its Gemini models. These sophisticated AI systems enable earlier and more precise detection of ads that violate policies. Remarkably, over 99% of such ads were identified and blocked before reaching users last year.
This data indicates a transformative shift in Google's approach to ad enforcement. While the number of problematic ads being intercepted has risen, the suspension of advertiser accounts has decreased. This trend suggests a strategic pivot from outright banning bad actors to a more nuanced method of evaluating and blocking individual ads.
Moreover, Google has observed that the increase in blocked ads correlates with the rise of generative AI misuse by scammers, who produce deceptive content at scale. The Gemini models play a crucial role in identifying patterns across extensive campaigns, allowing for earlier intervention.
This evolution aligns with Google's broader initiative to integrate its Gemini models into various core products and infrastructure, particularly in advertising. The company is increasingly leveraging AI to automate campaign creation, detect policy violations, and respond to emerging threats in real-time.
In the U.S. alone, Google removed over 1.7 billion ads and suspended approximately 3.3 million advertiser accounts in 2025. Common violations included ad network abuse and misrepresentation. In India, Google's largest user base, the company blocked nearly 483.7 million ads, almost double the previous year's figures, while account suspensions fell significantly.
Keerat Sharma, Google's VP and general manager of ads privacy and safety, emphasized the shift towards more targeted, AI-driven enforcement. This approach has notably reduced incorrect suspensions by 80% year-over-year. Sharma highlighted the importance of advertiser verification processes, which require businesses to confirm their identity before launching ads, contributing to the decline in suspensions.
As Google continues to refine its defenses and adapt to the tactics of bad actors, the company remains committed to intercepting harmful ads as early as possible in the advertising pipeline. This proactive stance not only enhances user safety but also sets a precedent for the future of digital advertising.