A recent study pitted artificial intelligence against human fraudsters in the cunning world of text-based romance scams. The results were eye-opening: an AI chatbot not only matched but often outperformed its human counterparts in building trust with potential victims.

The researchers from Amrita Vishwa Vidyapeetham, Foscari University of Venice, the University of Melbourne, and Ben Gurion University of the Negev used a form of fraud known as ‘pig butchering,’ where scammers build a relationship before shifting to fake cryptocurrency investments. In their experiment, AI chatbots convinced nearly 50% of test subjects to download an app or play a game, compared to just under 20% for human scammers.

The trust-building phase, which can last months in real-world scams, was the standout area where the AI excelled. Test subjects rated their level of trust with the chatbots significantly higher than with the humans. This suggests that as AI technology advances, fraudsters may find it easier to deceive potential victims over extended periods.

The findings highlight a concerning trend: as AI capabilities continue to improve, we might need more sophisticated tools and techniques to combat online scams. The future of scamming could very well be an automated one, making detection and prevention increasingly challenging.