Microsoft is utilizing AI to stop phone scammers from tricking users. A new service known as Azure Operator Call Protection will reportedly look out for suspicious phone calls and then warn users accordingly.
Microsoft Is Utilizing AI to Stop Phone Scammers
Microsoft is addressing the issue of suspicious calls that users actually answer with its new service called Azure Operator Call Protection. This service analyzes conversations in real-time and can notify the user if the caller appears to be suspicious. Microsoft has been testing the program with BT Group and is showcasing its functionality at the Mobile World Congress conference in Barcelona.
This announcement comes amid the persistent problem of spam calls. According to a study by voice security platform Hiya, the average phone user receives about 14 spam calls per month out of 98 billion phone calls analyzed worldwide. The Federal Communications Commission has recently taken action against robocalls by declaring scam calls made using AI-generated voices to be illegal.
Azure Operator Call Protection
Azure Operator Call Protection is designed for mobile carriers to offer to their subscribers as an option. It leverages AI to detect signals of potential fraudulence during a conversation. These signals could include language prompting the recipient to disclose sensitive information over the phone, explains Shawn Hakl, vice president of 5G strategy for Microsoft’s Azure for Operators program.
Hakl emphasizes that the tool reinforces best practices and aims to combat common scams, such as impersonations of Amazon, insurance providers, credit card companies, and attempts to obtain Medicare information. The AI model will evolve to adapt to new threats over time.
The current version of the service will interrupt the call to alert the user if it detects potential fraud. The user then has the option to end the call or inquire further about why it was flagged. The system does not automatically terminate the call; the user retains control over the decision-making process.
How the Service Works
The service is opt-in, requiring users’ consent, and data from calls is not stored or used to train Microsoft’s AI models. Once the call concludes and the user responds to the recommendation or makes additional queries, the data is discarded.
While Microsoft is collaborating with BT Group to trial the technology, there is no indication of when it will be available in the market. Microsoft’s initiative aligns with broader efforts to combat phone scams, including AT&T’s addition of logos to legitimate business calls and a virtual hackathon announced by the White House focused on developing AI technologies to identify unwanted spam and robocalls.
The Future of Azure Operator Call Protection
Looking ahead, the feature may expand beyond voice calls to include text-based communications. However, the focus is currently on evaluating voice-based interactions.
