Google Rolls Out Fake Call Detection to Fight AI Deepfake Scams

Google is rolling out a new AI‑powered fake call detection feature across Android devices to help shield users from increasingly sophisticated impersonation scams that use generative AI to mimic the voices of trusted contacts, financial institutions, or officials. The tool, launching in the Phone by Google app this month on devices running Android 12 and up, starting with Google Pixel models, marks a proactive defense against fraudsters exploiting AI voice‑cloning techniques.

The threat environment for voice scams has grown substantially with the availability of deepfake tools capable of producing highly realistic audio. According to data from INTERPOL’s 2026 Global Financial Fraud Threat Assessment, impersonation fraud contributed significantly to more than $400 billion in global losses, while U.S. Federal Trade Commission figures show reported impersonation scam losses of $2.95 billion in 2024 alone figures that underline the urgency for new defenses.

Industry‑First Digital Handshake Model

Google’s fake call detection works by leveraging Rich Communication Services (RCS), a next‑generation messaging standard that enables a secure “digital handshake” between Android devices using the Phone by Google app. When a contact initiates a call, the caller’s device sends a silent real‑time confirmation signal encrypted end‑to‑end to the recipient’s phone. If the signal is missing because the call is spoofed or routed through a scammer’s system, the phone displays a warning urging the user to hang up.

The feature builds on Google’s verified call protections, which already warn users about potential impersonation attempts by malicious actors posing as financial institutions. The new capability specifically targets deepfake voice scams, where attackers use machine‑generated voices that mimic tone, cadence, and speech patterns of a known person to manipulate victims into divulging sensitive information or transferring funds.

How It Helps Consumers and What It Requires

For the feature to function effectively, both the caller and recipient must be using Phone by Google, and the necessary Google Messages and Google Contacts apps must be installed to support RCS verification. When these conditions are met, the silent confirmation happens automatically behind the scenes, requiring no action from the user until a potential spoofed call is detected.

Android’s approach represents one of the earliest mobile‑level implementations of AI‑driven voice impersonation detection aimed at real‑time alerts. The Verge notes that such protections cover scams that accounted for over $893 million in U.S. losses in 2025, illustrating why consumers and security advocates have urged platform‑level defenses against emerging AI threats.

A Broader Tech Security Arms Race

While this feature enhances user protection, experts caution that deepfake detection remains an evolving challenge. AI voice‑cloning tools continue to improve rapidly, and scammers may adapt by targeting devices or services not using RCS verification. Security researchers emphasize the importance of layered defenses, including user awareness, multi‑factor authentication, and ongoing improvements to generative‑AI countermeasures at both the platform and application level.

Google’s rollout reflects a broader trend in which major tech companies are deploying AI both for malicious impersonation tactics and for defenses against them. By integrating verification protocols into the core phone experience, Google’s fake call detection aims to lessen the risk of financially and emotionally costly scams that exploit trusted identities with increasing realism.