U.S. Postal Inspection Service warns of new AI-powered scams targeting public

Source Domain: www.wmar2news.com

BALTIMORE — The U.S. Postal Inspection Service is warning the public about a new wave of scams powered by artificial intelligence.

Scammers are using AI tools to make their schemes more realistic and harder to spot.

“So, what we’re seeing more and more now, the scams are the same, but you’re seeing it become more realistic, whether it’s voice cloning or image cloning and just being able to utilize AI to reach the public more easily,” Eric Shen said.

Shen, a United States Postal Inspector with 20 years experience, noted that scam attempts are prevalent across various platforms.

“Text messages, emails, some phone calls as well,” Shen said.

The agency, which has been around for more than 200 years, is involved because many of these scams still connect back to the mail.

“Whether it’s money being moved through the mail or continuous communication that’s being done through the mail,” Shen said.

In one lottery fraud scheme, a Jamaican national in Maryland was recently sentenced to about 3 years in prison and ordered to pay nearly $4 million in restitution. Investigators say victims were told to pay taxes and fees before claiming fake winnings, sending money through the mail and losing their life savings.

Shen has a personal connection to the issue, as a family member fell for an investment scam.

“I have family members that have become victims and yeah, I’ve been an inspector for over 20 years. And that’s the reason why I’m still an inspector. I love this job and I love what I’m able to do to help the public,” Shen said.

Investment scams had the highest reported losses to the FBI Internet Crime Complaint Center in 2024, totaling $6.5 billion. This is a fraction of the $16.6 billion reportedly lost in cyber crime in 2024.

To combat the issue, the U.S. Postal Inspection Service is asking the public to “spot the bot.” Tell-tale signs of AI-generated images include background inconsistencies, extra fingers on hands, and incorrect scaling of objects like trucks. These images often…

Source