cyber.hendryadrian.com
Security News · helpnetsecurity.com

Google Reports Blocking 602 Million Scam Ads as Gemini Supports Detection

· Original article ↗

Summary

Google says Gemini-powered systems analyze hundreds of billions of signals to detect deceptive ads. In 2025, it removed or blocked 8.3 billion ads, including 602 million associated with scams, and acted on over four times as many user reports as in 2024.

Key points

  • Google says Gemini models analyze account age, behavioral cues and campaign patterns to detect deceptive ads based on intent.
  • Google reported removing or blocking 8.3 billion ads in 2025, including 602 million associated with scams, and suspending 24.9 million accounts, including 4 million associated with scams.
  • The company says improved user-feedback processing helped teams act on more than four times as many reports in 2025 as in the previous year.
  • Google reported removing 1.7 billion ads in the US, 675.7 million in the UK and 1.6 billion in the EU.
  • Google notes that some malicious ads still get through its automated checks.

Article Details

Event Type
Google's reported use of Gemini AI models to detect and block deceptive ads, including scam-related ads, across its advertising network.
Impact
Google reported that in 2025 it blocked or removed over 8.3 billion ads and suspended 24.9 million accounts, including 602 million ads and 4 million accounts associated with scams. It said Gemini's intent-based analysis and improved processing of user reports helped teams act on more than four times as many reports as the previous year. These figures and performance claims are attributed to Google.

People

Keerat SharmaGoogle's VP and GM of Ads Privacy and Safety; described how Google's teams use Gemini models to detect and block deceptive ads.

Vendors

GoogleGoogle says its security teams use Gemini AI models around the clock to detect and block deceptive ads in real time by analyzing hundreds of billions of signals.

Products

GeminiGoogle says its security teams use Gemini AI models around the clock to detect and block deceptive ads in real time by analyzing hundreds of billions of signals.

Countries

United KingdomRegional takedowns included 1.7 billion ads in the US, 675.7 million in the UK, and 1.6 billion in the EU.United StatesRegional takedowns included 1.7 billion ads in the US, 675.7 million in the UK, and 1.6 billion in the EU.

Related Articles