[Capture News] Voice phishing scammers caught wrong in AI chatbot...A counterattack that has become a "terrible nightmare."

Daisy, a grandmother's artificial intelligence chatbot released by a British news agency.

Daisy's client is none other than a voice phishing agent.

[Daisy / Artificial intelligence chatbot: (It's almost an hour). Hell!) Time goes by so fast.]

The goal is to ask again as if you don't understand without giving personal information, recite personal catch, and hold voice phishing as long as possible.

I talk to a scammer in real time without human intervention, and I actually called for 40 minutes, the news agency said.

[Daisy / Artificial Intelligence Chatbot] Hello, crooks. I will be your nightmare.]

Domestic telecommunication companies have also released technology to detect voice phishing with artificial intelligence.

"Cha Eun-Woo, listen carefully. I just hacked your phone.It's "
"."

If suspicious circumstances are detected during the call, a 'Voice phishing risk' notification will be displayed.

"Voice phishing suspect call."

Artificial intelligence, which analyzed and learned 20,000 voices of existing voice phishing criminals, was used.

[Kim Jung-ju / AI Privacy Team Leader, Korea Internet & Security Agency: Voice phishing scammers often tell crime victims personal information such as their social security numbers, names, and addresses to make them look like they are in real life. Although personal information was removed as much as possible, the pseudonym information system was used under the Personal Information Protection Act so that it could also be used as artificial intelligence learning data.]

As voice phishing techniques evolve rapidly, experts advise that fraudulent data should be updated periodically and new prevention measures should be prepared.

I'm YTN's Jang Ayoung.

Edit Video|Interest
subtitle news|This line

#YTN Caption News