Google is implementing a new Chrome security facility that uses the underlying ‘Gemini Nano’ Large-Language Model (LLM) to detect and block the tech support scam when browsing the web.
Tech support scams are malicious websites that users think that their computer has a virus infection or other problem. These alerts will be shown as a full-screen browser window or will display additional pop-ups, making them difficult to close.
The goal is to persuade the victim to call a listed number to help unnecessary remote support membership or help to get remote access to equipment, leading to financial loss or data theft.
Source: Bleepingcomputer
Google Chrome 126 AI for AI directly for fast, privacy-centered assistance within the browser.
The new anti-scam system of Chrome, integrated into the ‘enhanced protection’ of the browser, analyzes the web pages in real time to detect scam signals such as fake virus alerts or full-screen lockouts, which are hallmarks of take support scams.
This analysis is offline, locally on the user’s device using Gemini Nano. When there is a positive match, the data (LLM output + site metadata) is sent to ‘Google safe browsing’ for more intensive evaluation.
If malicious intentions are confirmed, Chrome will display a warning message informing the user of risk.
Source: Google
Google says that this feature respects the privacy of the users and has only the minimum performance effect, although several details were not given in the announcement.
“It is all done in a way that preserves performance and privacy,” Google announced,
“In addition to ensuring that the LLM is only rarely triggers and run locally on the device, we carefully manage the consumption of the resource by considering the number of tokens used, running the process incomprehensible to avoid disrupting browser activity, and implement the throttleing and cota operational system to limit the GPU use.”
The AI-operated security facility will be applied to Chrome 137, which will be determined for release next week, and it will be enabled by default on all users that upgrade to the latest version and opt for ‘enhanced security’ within the safe browsing settings of the browser.
open Chrome Settings> Privacy and Safety> Protection> Protection Safety To enable it.
Google said that it is a plan to expand the system in future release and make it capable of detecting other scam types, such as fake package delivery or toll notice. In addition, Chrome for Android will get this facility for some time in 2025.
Google’s new anti-scam feature was introduced by Microsoft earlier this year, which uses a specially trained machine learning model to detect and block the user target scams.
Based on the analysis of 14M malicious tasks, search for the top 10 MITERAT & CK techniques behind the 93% attacks and how to defend them against them.
