Amazon Built a Scam Checker. Here's What That Tells Us.

Amazon's answer is a new Alexa for Shopping feature that lets U.S. customers describe any message, email, phone call, or text and check it against Amazon's catalog of billions of sent communications. Suspicious submissions get reported automatically. It's fast, it's useful, and it's also a quiet admission that impersonation fraud has grown large enough to require a dedicated internal tool at one of the world's largest companies.

The FTC consistently ranks Amazon among the most impersonated brands in the United States. In 2023, Americans reported losing $10 billion to fraud, with impersonation scams making up the largest category. Amazon's 360,000 annual contacts are only the people who thought to call. The ones who didn't, and clicked anyway, never show up in that count.

Why the Brand Works for Scammers

Almost everyone has an Amazon account. Almost everyone gets real Amazon emails. The fake ones arrive with familiar logos, order numbers, and urgent language about a charge or a package. The calls go further: a professional-sounding voice says your account has been compromised, an unauthorized purchase went through, your Prime membership is about to be suspended. Voice cloning has sharpened the edge. Generating a convincing synthetic voice now requires as little as 20-30 seconds of audio.

The Alexa feature answers one specific question: "Did this come from us?" It works because Amazon controls both sides. They sent the messages, so they can check whether the one you received matches their records. That logic does not extend to the scam that causes the most financial damage: a caller who sounds exactly like someone you love.

In 2023, a Canadian family lost $21,000 after receiving a call from someone who sounded exactly like their son, claiming he had been in a car accident and needed bail money. The voice was a clone. The son was at home, unaware any of this was happening. The script is old; the grandparent scam has existed for decades. What changed is that the voice no longer sounds like a stranger trying to sound like your grandson. It sounds like your grandson.

When a caller says "It's me, I'm in trouble, I need money," there is no corporate database to check against. There is just a voice that sounds right and a story that demands an immediate answer.

A Check That Moves Faster Than a Clone

Trust Onion was built for exactly this gap. The app gives families three rotating codewords calculated locally on each person's phone. The words change every 60 seconds. When a call comes in and something feels wrong, one question settles it: "What are the words?" A real family member opens the app and reads them back. A scammer with a cloned voice and a convincing story cannot answer, because the words live only on your family's phones and the ones from a minute ago have already expired.

A failed callback does not prove the original caller was legitimate, so the codewords matter: the caller must still know your family's private Trust Onion Three Words. If something feels dangerous, alert the authorities.

Amazon's tool asks, "Is this really you?" for messages. The codewords ask the same question on a phone call, where you have seconds and no logo to examine.

When a company with Amazon's resources decides impersonation is serious enough to build a dedicated feature, that tells you something about the scale of what is happening. The Alexa tool helps millions of customers. It cannot help when the voice on the line belongs, or seems to belong, to someone you would do anything for.

A controlled voice-cloning safety demonstration

Try our tool to see just how easy voice cloning is. This controlled safety demonstration exists only to show how easy voice cloning is and to make the risk tangible. It is not a real incident or proof of anyone's identity. Your submitted sample is handled transiently to generate the fixed demonstration audio. Trust Onion does not identify you or the person from the sample. Trust Onion does not store the submitted recording, clone, or generated audio.