Alphabet Inc.'s (NASDAQ: GOOGL) (NASDAQ: GOOG) Google has launched an aggressive new defense against internet scams, deploying its on-device Gemini AI model to detect and block fraudulent websites in real-time, significantly expanding its security capabilities across Chrome, Search and Android operating systems.
What Happened: Google is now using Gemini Nano, a version of its artificial intelligence model that runs directly on users' devices, to scan webpages instantly when clicked to identify potential threats, the company announced Thursday.
This technological advancement specifically targets "tech support" scams - deceptive pop-ups claiming device infections that prompt users to download fake security software or call fraudulent support numbers.
"Fighting scammers has always been an evolution game," said Phiroze Parakh, senior director of engineering for Google Search, in an interview with CNN, noting that while bad actors now have new AI tools, so do security teams.
Why It Matters: The security enhancements come amid rising global concerns over online fraud, with consumers worldwide losing more than $1 trillion to scams last year, according to the Global Anti-Scam Alliance, the CNN report notes.
Since implementing AI-powered anti-scam systems three years ago, Google now blocks 20 times more problematic pages, removing hundreds of millions of scam search results daily. The company reports it has decreased scam attacks in airline-related searches by 80%.
This security initiative builds on Google's broader AI security strategy that includes features like Theft Detection Lock, Offline Device Lock and Remote Lock for Android devices.
Meanwhile, Google CEO Sundar Pichai revealed ongoing negotiations with Apple Inc. (NASDAQ: AAPL) to integrate Gemini AI technology into iPhones by mid-2025, potentially extending these security capabilities to the iOS ecosystem.