U.S. Postal Service Warns of AI Scams
The U.S. Postal Inspection Service is warning the public that criminals are using artificial intelligence to make traditional scams more convincing. The warning, issued during National Consumer Protection Week, highlights how AI can make fraudulent schemes appear more realistic and harder to detect.
Criminals are leveraging artificial intelligence to enhance long-running fraud schemes, including romance scams, investment scams, and tech support cons. AI-generated photos and voice clones that mimic real people are used to build trust with victims over extended periods, making the scams more believable and harder to detect. One of the most concerning techniques is voice cloning, where scammers can replicate a person's voice from just a few seconds of audio. This has led to a rise in "grandparent scams," where a cloned voice of a relative is used to make a fake, urgent plea for financial help. These AI-generated calls create a sense of distress that can be very convincing. Phishing attacks have also become more sophisticated, moving beyond emails with obvious spelling errors. AI, particularly large language models, can now generate highly convincing scam messages and websites that mimic the tone and style of trusted brands, making them difficult to distinguish from genuine communications. Deepfake technology, which uses AI to create realistic but fake videos, is another tool being used to deceive. Scammers have used deepfakes to impersonate CEOs in video calls to authorize fraudulent financial transfers and have created fake celebrity endorsements for cryptocurrency scams. The U.S. Postal Inspection Service advises consumers to be wary of unexpected requests for money, even if they appear to come from a known person. They recommend hanging up and calling the person back on a known number to verify the situation. For suspicious emails or texts, the advice is to never click on links and to independently verify the sender through official websites or phone numbers. Telltale signs of an AI-driven scam can include a person's social media profile being very new or having few friends, inconsistencies in video or audio, and pressure to move communications to a private channel like WhatsApp or Telegram. Authorities also warn against making payments via wire transfer, gift cards, or cryptocurrency, as these methods are difficult to trace and recover. If a scam is suspected, it should be reported to the U.S. Postal Inspection Service, the Federal Trade Commission (FTC), and the FBI's Internet Crime Complaint Center (IC3). Victims should also immediately contact their bank to try and stop any financial transactions.
