Google wipes out 602 million scam ads with Gemini on duty

Google reports that its Gemini AI models are used to detect and block malvertising and deceptive ads at scale, analyzing hundreds of billions of signals (account age, behavioral cues, campaign patterns) to stop threats. In 2025 the company says it blocked or removed over 8.3 billion ads and suspended 24.9 million accounts, noting attackers increasingly use generative AI to create deceptive ads that can deliver malware or phishing pages.