Chrome Enhances Scam Protection with Gemini AI and ClickFix Defense
**Google Chrome** is rolling out significant updates to its Enhanced Protection in Safe Browsing, leveraging **Gemini** AI to detect deceptive websites and introducing a novel defense against **ClickFix** attacks. These advancements aim to bolster user security by offering more robust real-time threat detection and proactive warnings against sophisticated social engineering tactics.
**Google Chrome** is stepping up its scam prevention game, integrating advanced **Gemini** AI capabilities to identify deceptive websites and warn users against suspicious commands often seen in **ClickFix** attacks. These enhancements are part of **Safe Browsing's** Enhanced Protection, which **Google** claims makes users ten times safer from phishing, social engineering, and other online scams compared to its Standard Protection.
A new "suspicious site" warning will also be introduced for pages that exhibit deceptive patterns but haven't yet met the threshold for a full-page red warning. This tailored approach allows **Chrome** to expand its protective coverage effectively, providing more timely alerts to users.
### Gemini AI Moves to Server-Side Scam Detection
Previously, **Chrome** utilized an on-device **Gemini** model for detecting tech support scams. The latest update shifts to a more powerful server-side model for Enhanced Protection users. This model analyzes various signals, including the brand represented on a page, its intent, and its likelihood of being a scam. This analysis is then combined with domain reputation and other security signals by **Safe Browsing**.
To safeguard user privacy, **Chrome** minimizes data sharing with the server, only analyzing page text when high-confidence scam indicators are present. Any page content sent off-device is processed ephemerally, meaning it is never logged, retained, or used for model training.
### New Defense Against ClickFix Commands
Enhanced Protection now includes the ability to detect when a user copies a suspicious command associated with a **ClickFix** scam. These scams typically trick users into executing malicious PowerShell or Command Prompt commands themselves.
If **Chrome** suspects a **ClickFix** script, it automatically removes personally identifiable information from the clipboard payload and sends it to **Safe Browsing** for analysis. Should the analysis deem the content dangerous, **Chrome** will display a warning, advising the user before they proceed with the potentially harmful action.