Google’s annual ads safety report released Thursday detailed how the company’s artificial intelligence blocked over 99% of policy-violating advertisements in 2025, part of a broader push to defend against AI-generated spam and scams. The company blocked or removed 8.3 billion ads total last year, up from 5.1 billion in 2024, with 602 million of those carrying scam-related violations, according to the report.
The acceleration of both scam-generation and scam-detection represents a turning point in online advertising security, where artificial intelligence tools serve both sides of an escalating arms race that experts predict will intensify.
The Scope of the Problem
The FBI’s Internet Crime Report documented more than 22,000 complaints reporting AI-related scams in 2025, with total losses exceeding $893 million. The rise of generative AI tools has exacerbated the perennial problem of online scams.
“It’s not that this is a new problem. It is an old problem, supercharged,” said Nate Elliott, a principal analyst at Emarketer. “The biggest difference is the speed and the scale that AI offers both the good actors and the bad actors.”
Google’s Gemini-Powered Defense
Google deployed its generative AI system, Gemini, to detect and block violations before ads reach audiences. Gemini analyzes hundreds of billions of signals — including account age, behavioral cues and campaign patterns — to determine whether an advertiser’s intent is legitimate or potentially malicious, according to Keerat Sharma, Google’s vice president and general manager of ads privacy and safety.
The speed of analysis improved dramatically. Evaluating digital assets in ads once required anywhere from a few seconds to minutes or longer, but Gemini now completes the same analysis in milliseconds. “That allows us to stop things right at the front door,” Sharma said.
Google suspended 24.9 million advertiser accounts last year, with more than 4 million of those suspensions tied to scam-related activity. Gemini helped reduce incorrect advertiser suspensions by 80%, allowing legitimate businesses’ ads to remain online, the report said.
Market Shifts and the Competition Ahead
Google generated more than $200 billion in net worldwide ad revenues last year, according to data from Emarketer. But the research firm predicts Meta will outperform Google in ad revenue in 2026.
A Google spokesperson said the company doesn’t report specifically how many ads it blocks that were AI-generated, because its enforcement approach targets policy violations regardless of how the ad was created. The spokesperson noted that many AI-generated ads comply with Google’s policies and come from legitimate businesses.
Experts anticipate the contest between AI-powered scams and AI-powered defenses will intensify as the technology advances. “We’re already close, but it’s going to be heading even more to where it’s just AI versus AI,” said Matt Seitz, director of the AI Hub at the University of Wisconsin-Madison. “The volume of this problem is so large that it can’t be managed directly through humans.”