Google has announced the deployment of AI-powered fraud detection capabilities to protect Android devices users and their personal information.
“These features are specifically targeted at conversation scams, which often seem harmless at first before they evolve into harmful situations,” Google said. “And more phone sales scammers use spoofing techniques to hide their actual numbers and pretend to be a reliable company.”
The company partners with financial institutions to better understand the nature of the fraud customers encounter, allowing them to devise an AI model that can flag suspicious patterns in the course of the conversation and provide real-time warnings without sacrificing user privacy.
These models run entirely on the device and alert users if there is a possibility of fraud. The user then has the option to reject or report and block senders. The setting is enabled by default and only applies to conversations using phone numbers that are not in the device’s contact list.
Tech Giant also highlighted that if a user’s conversation remains private and chooses to report the chat as spam, recent messages with senders will be shared with Google and carriers.
The feature was first released in the US, UK and Canada in English, with wider expansion planned at a later date.
Additionally, similar fraud detection for phone calls has been extended to all English-speaking pixel 9+ users in the US. This feature was first introduced in November 2024.
It is worth pointing out that while this feature is turned off to provide user control by default, call audio is processed short-lived and not used during contact calls.
“If enabled, fraud detection calls the feature is on to notify participants at startup and to notify them,” Google said. “You can turn off fraud detection at any time during individual calls or during all future calls.”
The development takes place a few weeks after Google revealed that it uses an enhanced protected mode of secure browsing in its web browser.
“When you turn on enhanced protections, much of the additional protections you receive comes from advanced AI and machine learning models designed to find dangerous URLs engaged in known phishing, social engineering and fraud techniques,” the company said.
“Safe Browsing’s enhanced protection model can identify URLs designed to look similar to trusted domains. It also uses advanced AI and machine learning technologies to protect against dangerous downloads.”
Source link
