Britain Develops ‘AI Granny’ to Combat Phone Scammers

The British telecommunications company O2 has developed an ‘AI Granny’ to tackle phone fraud. Her primary role is to feign confusion, endlessly ask questions, and waste time.

The neural network, named Daisy, can answer suspicious calls around the clock and engage in realistic conversations with scammers. Initial tests showed she can keep criminals on the line for up to 40 minutes.

The ‘AI Granny’ spins elaborate tales about her family, talks about her love of knitting, and provides false personal information such as fictitious bank details. She makes scammers believe they are deceiving a real person.

Daisy occupies criminals’ time, preventing them from targeting real victims. She has also analyzed common scam tactics, enabling people to better protect themselves.

O2’s initiative emerged following a survey indicating that 67% of British adults are concerned about falling victim to scams, with 22% encountering them weekly.

“Our new team member in the fight against fraud, Daisy, is a game-changer, outsmarting criminals in their own cruel game by simply keeping them on the line,” commented Murray McKenzie, Director of Fraud Prevention at Virgin Media O2.

When a user receives a suspicious call, they can redirect it to the number 7726. From there, Daisy conducts the conversation, generating responses in real time.

The system transforms the caller’s speech into text, processes it with artificial intelligence to generate a response, and converts it back into voice. This creates natural conversations.

In August, Binance reported using AI against fraudsters, preventing the theft of over $2.4 billion in user funds.