Spamming the spammers: Giving spam callers a taste of their own medicine

Spam calls have become a scourge, with billions of phone calls, a good number of them being robot calls, causing annoyance by wasting time and financial losses to millions of people across the world. Using cheap and free telecommunication tools, spammers make billions of phone calls to people to dupe them into deals.

They play on human emotions and fears, making them transfer money. How about beating these spam scammers in their own game, giving them a taste of their own medicine?

Cybersecurity experts have come out with an artificial intelligence-based solution that can waste the time of the callers. It is a kind of scamming the scammers and can become a big deterrent to the spam callers.

Also read: WhatsApp rolled out fixes to tackle international spam calls

A team from Macquarie University Cyber Security Hub analysed scam phone calls and identified the social engineering techniques that scammers use on their victims. They deployed machine learning techniques and Natural Language Processing (NLP) to identify typical scam ‘scripts’. They then trained chatbots on a dataset of real-world scam conversations from recordings of scam calls to transcripts of scam emails, and chat logs from social media platforms so the bot can generate its own conversations resembling those of real-world scam calls.

The system, developed by the Australian university team, will deploy convincing voice clones to launch conversations with real scammers. Named after the Greek Goddess of Deception, the solution Apate promises to make a dent into the revenues of the spam call industry by cutting into their time and capacity to spam the gullible.

“Phone scams are run by organised crime groups and currently only a tiny fraction of criminals are caught, and the money is rarely recovered,” Professor Dali Kaafar, Executive Director of Macquarie University’s Cyber Security Hub, says.

He said his team now has patents pending for this highly-effective technology.

Also read: TRAI tightens guidelines to control unwanted calls, messages

Technologies like voice-over-internet protocol (VOIP) make it easy and cheap for cyber-criminals to mask their location, pretending to call from any number. It is hard and expensive to update the telecommunications infrastructure and protocols to improve the authentication of the calls.

“Financially, it’s a high-gain, low-cost ratio for scammers, the practice is very lucrative and a relatively low-risk criminal activity - and it’s pretty hard for victims to recover this money,” he said in a statement.

“Our model ties them up, wastes their time, and reduces the number of successful scams. We can disrupt their business model and make it much harder for them to make money,” he said.

Also read: How Generative AI can strengthen your cyber defence arsenal

Understanding spammers

The tools allowed them to develop AI agents that are capable of fluent speech and can adopt a particular persona and stay on track in a conversation, being convincingly consistent in their responses.

“The conversational AI bots we have developed can fool scammers into thinking they are talking to viable scam victims, so they spend time attempting to scam the bots,” Kaafar says.

These bots can be trained in any language or accent and because phone scams are a global challenge, this technology can be deployed anywhere in the world.

COMMENT NOW