For over ten years, Google has leveraged advancements in AI to shield you from online scams, where malicious actors deceive users to access money, personal data, or both. Today, we’re publishing a new report on how we combat scams in Search and revealing new AI-driven methods to protect you across Search, Chrome, and Android. Stopping scams in Search with AI-powered defenses AI enables us to detect and block hundreds of millions of scam-related results daily in Search. Our report, Fighting Scams in Search, highlights how investments in AI-based scam detection systems, alongside enhanced classifiers, have allowed us to identify 20 times more scam pages. These improvements guarantee that the search results you receive are legitimate and shield you from harmful sites seeking to steal your sensitive information. AI advancements have strengthened our scam-fighting tools, empowering us to analyze vast amounts of web text, spot coordinated scam campaigns, and identify new threats—keeping us steps ahead to protect you during Search. For instance, we’ve noticed a sharp rise in scammers impersonating airline customer service to trick vulnerable users. Thanks to these efforts, we’ve reduced such scams in Search by over 80%, significantly lowering the chances of you calling a fraudulent phone number. Supercharging Safe Browsing in Chrome with Gemini Nano Chrome’s Safe Browsing Enhanced Protection mode offers the highest level of security and makes users twice as secure against phishing and scams compared to Standard Protection. Now, we’re integrating Gemini Nano, our on-device large language model (LLM) for desktop, to provide Enhanced Protection users with an extra defense layer against online scams.

This on-device method offers immediate evaluation of risky sites and protection even from previously unseen scams. Gemini Nano’s LLM is ideal due to its capacity to distill the intricate and diverse nature of websites, enabling quicker adaptation to evolving scam techniques. Currently, this AI-driven approach safeguards users from remote tech support scams, one of today’s most prevalent online threats. We aim to extend this protection to Android devices and cover more scam types going forward. Fighting scams, spam, and unwanted notifications Sometimes, the danger from scammy sites extends beyond just the website. If you have enabled website notifications, malicious sites may bombard you with scammy alerts. To help you stay protected from deceptive, spammy, or misleading notifications, we are introducing new AI-powered warnings for Chrome on Android. When Chrome’s on-device machine learning model detects a suspicious notification, you’ll receive a warning offering the choice to unsubscribe or view the blocked content. If you believe a warning was triggered in error, you can allow future notifications from that site.