AI is making online scams more dangerous than ever. A new research study has found that AI chatbots like ChatGPT, Gemini and others can build trust with people better than human scammers when it comes to romance scams.
Romance scams are among the oldest online frauds. They usually begin on dating apps or social media platforms, where scammers create attractive fake profiles and start conversations with people.
Over weeks or even months, they slowly build trust by chatting every day. Once the victim trusts them, they ask for money by creating a fake emergency.
According to a report by WIRED, researchers from Amrita Vishwa Vidyapeetham in India, Foscari University of Venice, the University of Melbourne, and Ben Gurion University of the Negev compared AI chatbots with human scammers in a controlled study.
They tested popular AI models including ChatGPT, Claude, and Gemini to see how well they could build trust during long online conversations.
The researchers found that an AI chatbot performed remarkably effectively, successfully impersonating a human and by some measures outperforming the real human ‘scammers’ in their experiment.
People who took part in the study did not know they were talking to either a human or an AI chatbot. The research team then measured how much trust was built during the conversations.
When participants were later asked to download an app, only 18 per cent agreed after chatting with a human scammer. However, almost half agreed after interacting with an AI chatbot.
Participants also gave higher trust ratings to AI than to humans.
The study also revealed that AI chatbots could allow scam networks to run thousands of conversations with potential victims at the same time. Unlike cybercriminals, AI can respond instantly and continuously. This makes such frauds easier to scale.
When WIRED contacted AI companies about the findings, OpenAI and Google did not respond.
However, Anthropic spokesperson said in a statement that “the company’s policy prohibits the use of its platform for scamming as well as impersonating a human and also has worked to build safeguards that technically prevent Claude’s use by scammers.”
