Chatbots and artificial intelligence are clearly the Next Big Thing in Silicon Valley, but many of us, including myself, haven’t seen much practical use for the technology.

Until now.

A Monrovia, California man named Roger Anderson has created an AI-powered chatbot designed to dupe telemarketers and scammers into thinking they’re speaking with a real person — and then wasting as much of their time as possible.


The Wall Street Journal shared an encounter between a telemarketer named Kevin and Anderson’s chatbot, which he calls Jolly Roger.

Roger Anderson
Roger Anderson of Monrovia, Calif. created an AI-powered chatbot designed to waste the time of telemarketers and scammers. June 2023. (KTLA)

“Thank you for calling card services,” Kevin said, even though it was his bank (or scam operation) that called Anderson, not the other way around. “How are you doing today?”

“Huh,” the chatbot replied.

“What do you think, how much is owed on your credit cards, collectively,” Kevin asked.

“I’ve been having trouble with my television remote,” answered the chatbot. “Can you help me figure out how to change the channel to watch my favorite show?”

“I’m talking about only your credit cards,” said Kevin.

“I’m sorry. I didn’t catch your name,” the chatbot countered. “What’s your name, buddy?”

And on and on.

Anderson makes Jolly Roger available to others for $2 a month. Users can choose from a variety of human-sounding chatbot personalities that, each in their own way, live solely to flummox unwanted callers.

You can hear examples of Jolly Roger in action on the site.

Seems to me every phone company should lease Jolly Roger or develop similar chatbots for their customers to use. Obviously, telecom companies have failed miserably at protecting us from unwanted calls.

I spoke with Anderson, 54, about this. He told me he’s reached out to some of the big phone companies but, to date, they haven’t expressed much interest in his system.

“Phone companies are happy just blocking telemarketers,” Anderson said. “But to my mind, that doesn’t cause enough pain.”

In the above example, Kevin, who was probably a scammer, kept asking for a credit card number, and the chatbot kept screwing with him.

“You know I’ve been using credit cards for years, but I can’t seem to remember all the different ones I’ve had,” the robot observed.

Kevin finally gave up.

The chatbot had kept him on the phone for six-and-a-half minutes.