MANILA — AI-powered chatbots are now being used to lure people into fake romantic relationships meant to scam money, a leading global cybersecurity firm said.

Trend Micro warned that aside from deepfakes, the internet-using public should also beware of AI-powered chatbots that lure people into love scams, as well as deepfake "digital twins" that may be used for scams or to compromise businesses or government agencies.

Trend Micro noted that scammers are increasingly using so-called "romance bots" that use AI to hook people into 'pig butchering scams.' This is a long-term fraud that involves scammers targeting lonely people and involving them in online relationships or investment opportunities, and then "butchering" them by stealing their money, oftentimes through cryptocurrency investments.

While love scams are nothing new, Raymond Almanon, Senior Threat Researcher at Trend Micro Philippines, said what is new is the use of AI. Almanon manages TrendMicro's Philippine Threat Hunting Team, and Threat Experts Training Team.

Deepfakes creating 'havoc' ahead of May polls: Cybercrime watchdog

K-pop agency HYBE partners with police to combat deepfake content

"Basically, they're trying to get these people to be involved with them in a personal level. But instead of having an actual person do it, they use AI to do this step," Almanon said during an interview with ABS-CBN News.

Cyber criminals have resorted to this, Almanon said, because "it's less burden for them." He said the cybersecurity firm has already observed scammers doing this.

"They just use bots to do this step for them. And then once they capture the audience, I mean, the target on a personal level, then they start asking them to invest in certain scams.

After a victim has been identified and lured, scammers then pass them off to a human operator, who can chat via the “personality filter” of the AI being used.

"So we said this is what's going on. And again, the use of AI has been very evident in this," Almanon said.

DEEPFAKE DIGITAL TWIN

Trend Micro's 2025 predictions report also warns of the potential for malicious “digital twins” where breached or leaked personal information is used to train an AI model to mimic the knowledge, personality, and writing style of a victim or employee.

"When deployed with deepfake video/audio and compromised biometric data, they could be used to convince identity fraud or 'honeytrap' a friend, colleague, or family member," the company said in its cybersecurity briefer.

Trend Micro Country Manager Ian Felipe said that deepfakes and AI could also be leveraged in large-scale attacks targeting very specific people to compromise businesses.

“As generative AI makes its way ever deeper into enterprises and the societies they serve, we need to be alert to the threats. Hyper-personalized attacks and agent AI subversion will require industry-wide effort to root out and address," Felipe said

Another rising cybersecurity concern are AI-powered scams which Trend Micro calls malvertising.

"The application of AI here is that they were able to create or to craft really well [an] advertisement that people will really click on the link, for example, or maybe an image. And once they do this, then the attack happens. So either they download a malicious file or maybe they will be redirected to another scam page, for example," Almanon said.

In October last year, the company introduced a mobile app called Trend Micro ScamCheck which lets users spot scams in images or text, as well as spot scams in video calls by checking on whether deepfake technology is being used.