Google is adding a dose of AI to the Chrome browser to protect users from tech support scams.
In a blog post, Google says a new version of Chrome for desktops this month will tap the “On Device” AI to flag and detect scams in real time.
These schemes often appear by popups that imitate security alerts to force the user to think that affects their computer with the virus. Sometimes, they show a phone number to call for help, but this is all scheme to do to pay the victims to pay fake tech support.
Although these scams have been going on for years, the scam behind them is permanently adapting to their tactics. Google noted that it could automatically extend popups to the full screen and disable the mouse input to create a “feeling of crisis”. Others have faced popups that play dangerous audio messages, and claim that the computer has been closed.
(Credit: PC Meg/Michael Kar)
In fact, there has been no infection. Popups are merely misusing the browser functions. To fight back, Google is using a large language model Gemini Nano, which was initially published on Pixel phones, to detect tech support scams on the browser.
Gemini Nano is small enough to run itself on the device, which enables him to flag threats in real time. The company said in a blog post, “We have noticed that the average malicious site has been available for less than 10 minutes, so on -device protection allows us to detect and prevent crawling attacks (through Google’s search engine).”
(Credit: Google)
Get our best stories!
Be safe with the latest greeting news and updates
Sign up for our Security Watch Newsletter for our most important privacy and security stories.
By clicking on the signup, you confirm that you are 16+ years old and agree to our use and privacy policy terms.
Thank you for signing up!
Your membership has been confirmed. Keep an eye on your inbox!
When Chrome goes to the malicious web page, the Gemini Nano Model will activate. Since tech support scams have mobilized some of the browser process, the Gemini Nano model can analyze the suspected malicious web page, including its intentions. The data is then sent to Google Google’s Safe Browsing Online Service for the “final decision”, which will determine whether Chrome should serve a warning that should be presented to the user to explain the potential threat.
From this point of view, there may be concerns about extracting Gemini Nano’s resources or attacking consumer privacy. But Google says it has come up with the guards to address such concerns.
The company explains, “In addition to ensuring that the LLM is just a bit dynamic and runs locally on the device, we consider the number of used tokens, to prevent the process from disrupting the browser activity, and to restrict the use of GPUs to prevent the use of throttle.
Suggested by our editors
It also looks like users can opt out. Google says the LLM Summerized Security Signals are only sent to its safe browsing service “for users who have chosen better protection of safe browsing in Chrome, protecting them from the risks that have not been seen before.”
(Credit: Google)
Meanwhile, users who have gone into chrome settings and will not send a big language data to Google to pick up a normal or no protection mode. But the company still notes: “Standard protection users will also benefit indirectly from this feature as we add newly discovered dangerous sites to block lists.”
Google’s security implementation on Gemini Nano arrived at the Chrome version 137 later this month. Microsoft has done so shortly after Microsoft introduced a similar AI -powered ability to detect and prevent “scarware” popups.
About Michael’s ear
Senior Reporter
