AI Is Getting Really Good at Messing With Cybercriminals
wired ·
Artificial intelligence is increasingly being deployed to counter cybercrime by tricking scammers into interacting with lifelike AI bots. The Australian company Apate has developed a system that diverts phone scammers onto calls with these bots, aiming to waste the criminals' time while collecting valuable intelligence on their operations. These bots are designed with diverse personalities and language skills to appear as realistic victims who express skepticism but remain engaged in conversation. By keeping scammers occupied for extended periods, the technology prevents them from targeting other potential victims using automated dialing tools. Beyond voice calls, similar AI strategies are being applied to infiltrate scam chat groups and respond to text messages. Researchers at ETH Zurich have also integrated large language models into honeypots to make virtual traps more convincing to AI-driven attackers. This shift represents a new era in anti-scam efforts, leveraging generative AI to disrupt criminal resources and gather data more effectively than traditional methods.
wired