Artificial intelligence is doing some incredible things right now. It can write code, generate art, and even help students study. But let’s not kid ourselves. It is also making life easier for scammers.
The U.S. Postal Inspection Service is using National Consumer Protection Week to sound the alarm. The law enforcement arm of the U.S. Postal Service says AI is making old school scams look a whole lot more believable. And that should concern all of us.
From March 1 through March 7, the agency is pushing out warnings online, on social media, on the radio, and even in airport ads in Washington, DC, Hollywood, Florida, and New York City. The message is clear. The tools criminals are using have evolved.
Chief Postal Inspector Gary Barksdale did not mince words. “From fake emails and texts to cloned voice messages, calls, and deepfake videos, today’s scammers have many tools at their disposal,” he said. “But if you’re alert and understand these criminals’ typical methods, you can stay one step ahead of them.”
That line stuck with me. Stay one step ahead.
Scammers are now leaning on AI generated photos, cloned voices, and deepfake video to impersonate real people. Romance scams. Investment schemes. Cryptocurrency traps. Fake tech support calls. Even so called emergency messages from “family members” who supposedly need money right now. With AI in the mix, these tricks can look and sound disturbingly authentic.
But here is the thing. While the tech is new, the psychology is not. The Postal Inspection Service outlined several red flags that still apply.
Start with social media. If someone contacts you out of nowhere and their profile feels thin, inconsistent, or freshly created, that is a problem. Too few friends. Photos that do not match the story. Comments that read oddly polished or robotic. Those are signs that something is off.
Then there are sketchy websites and email addresses. Tiny spelling mistakes in a domain name. A non secure site asking for login credentials. An email that looks official at first glance but does not match the real domain. That is not an accident. It is bait.
The agency also warns about overly polished “evidence.” Scammers can now generate slick looking PDFs, fake account statements, and convincing screenshots in minutes. Just because it looks professional does not mean it is legitimate.
Another classic move is trying to get you off a platform. If someone insists on switching to WhatsApp, Telegram, or a private email address, that is often about avoiding built in protections and moderation. When someone wants less oversight, you should want more.
And yes, pay attention to audio and video. Lip sync that feels slightly off. Strange lighting. Jerky movement. A voice that sounds close to someone you know but not quite right. AI cloning is impressive, but it is not perfect.
The Postal Inspection Service’s core advice is refreshingly simple. Ignore and delete messages that demand a quick decision. Be extremely wary of anyone asking you to send money via cash, wire transfer, or gift card. Once that money is gone, it is usually gone for good.
As someone who covers AI regularly, I think this is the uncomfortable side of the story that does not get enough attention. We love talking about productivity boosts and futuristic features. But the same technology that can generate a podcast voice in seconds can also mimic your spouse or your boss.
That does not mean we panic. It means we slow down. Verify identities through a known phone number. Check domains carefully. Do not click links in unsolicited messages. And never send money because someone pressures you in the moment.
AI may be getting smarter. That just means we have to get a little more skeptical.
Support independent tech journalism
NERDS.xyz is independently owned and operated. If you enjoy my coverage of Linux, AI, hardware, cybersecurity, and tech culture, consider supporting the site on Ko-fi.
Support NERDS.xyz