Google has begun rolling out its AI-powered Scam Detection feature for the Pixel Phone app, available exclusively to beta users in the US. Announced during Google I/O 2024 in May, this feature uses on-device AI to analyze audio from incoming calls and detect potential scams. Eligible devices include all Pixel models from the Pixel 6 (2021) onwards.
When Scam Detection identifies a suspicious call—such as someone posing as a bank representative to request an urgent money transfer—it issues warnings through audio, haptic alerts, and on-screen notifications. This functionality builds on Google’s existing security tools, such as Call Screen, which uses bots to vet calls before users engage.
The processing for this feature occurs entirely on the device to maintain privacy. Newer Pixels, like the Pixel 9 series, utilize the Gemini Nano chip, while older models rely on other advanced on-device machine learning systems.
As generative AI technology advances, both cybercriminals and security developers are leveraging it in the realm of digital security. Scammers are using AI to enhance fraudulent calls, while features like Scam Detection aim to stay one step ahead. Google plans to extend the functionality to other Android devices in the future, further bolstering its commitment to protecting users from evolving threats.
