Gemini blocks billions of ads

Google says Gemini now screens ads across its systems before they run, helping block or remove more than 8.3 billion ads in 2025. (blog.google) Google released those figures in its 2025 Ads Safety Report on April 16, 2026. The company said it also suspended 24.9 million advertiser accounts and stopped more than 99% of policy-violating ads before a user saw them. (blog.google) The company said 602 million of the blocked or removed ads were tied to scams, along with 4 million scam-linked accounts. Google also said advertiser identity verification remained a second layer of screening alongside automated detection. (blog.google) Online ad enforcement is the system that decides whether a campaign is allowed to run, much like an airport checkpoint deciding what gets through and what gets stopped. Google said Gemini is now used to analyze signals such as account behavior and payment patterns, so it can catch suspicious campaigns earlier in the review process. (blog.google) (emarketer.com) Google and outside trade coverage both framed the change as a shift from slower, rule-based moderation toward automated classification at very high volume. Search Engine Land reported Google said the upgrade also reduced mistaken suspensions of legitimate advertisers, while eMarketer said the tighter screening is meant to filter bad ads without sweeping in compliant campaigns. (searchengineland.com) (emarketer.com) The year-over-year numbers show how sharply the system changed. In its 2024 Ads Safety Report, published April 16, 2025, Google said it blocked 5.1 billion ads and suspended more than 39.2 million accounts in 2024. (blog.google) (services.google.com) That means Google blocked far more ads in 2025 while suspending fewer accounts, suggesting the company is getting more aggressive at stopping individual campaigns before moving to full account bans. TechCrunch described that pattern as Google “targeting bad ads over bad actors” as its artificial intelligence systems improved. (blog.google) (techcrunch.com) Google has tied the push to a rise in AI-assisted scams, including impersonation ads that mimic public figures or businesses. The Associated Press reported the company is using artificial intelligence to fight the same kind of AI-powered fraud that has spread across the web. (apnews.com) For advertisers, the practical change is earlier screening and less room for borderline creative to slip through review. For Google, the report turns ad policing into a speed-and-accuracy problem at internet scale, with Gemini now sitting much closer to the front gate. (emarketer.com) (blog.google)