How AI Is Changing Scams, and What Still Gives Them Away

A phone call in your grandson’s voice. A video of a celebrity recommending an investment. A perfectly written email from your “bank.” Artificial intelligence (AI) tools have made scams faster to produce and harder to spot by ear or eye. The good news: the scams themselves haven’t changed much. They still want the same things, and the same habits still stop them.

This article explains how criminals are using AI, why the old warning signs are fading, and what still gives a scam away. It was checked against Federal Trade Commission (FTC) and FBI guidance on November 9, 2026.

What AI changes for scammers

Voice cloning

The FTC warns that a scammer needs only a short clip of someone’s voice, which can come from a video posted online, plus a voice-cloning program, to make a call that sounds like a family member. These calls are often used in “family emergency” scams: a loved one in trouble who needs money fast. See our guide to AI voice-clone scam calls.

Polished writing in any language

Spelling mistakes and clumsy grammar used to be a reliable clue. The FBI says criminals now use generative AI to write convincing messages for social engineering, romance, investment and job scams, and to translate them so they read naturally. A well-written message is no longer a sign of a real one.

Fake photos, profiles and documents

AI can create realistic profile photos of people who don’t exist, along with fake IDs, badges and “proof” documents. The FBI notes these are used to build fake social media profiles and to back up romance and investment stories.

Deepfake video

Short fake videos of celebrities, business leaders or officials are used to promote investment schemes and giveaways. Some scammers also use real-time video tricks in video calls to “prove” they are who they say they are.

Scale and personalization

AI chatbots let one scammer hold many conversations at once and tailor messages using details scraped from public posts. That makes targeted phishing, which used to take time, cheap and fast. See How Scammers Get Your Information.

What AI doesn’t change

Whatever the technology, a scam still needs you to do something: send money, share a code, click a link or install an app. The FTC’s core warning signs still apply:

• Someone you trust, or claims to be. A familiar voice or face is no longer proof on its own.
• A problem or a prize. An arrest, an accident, a frozen account, a once-in-a-lifetime investment.
• Pressure to act now. Don’t hang up, don’t tell anyone, the deadline is today.
• A specific way to pay. Gift cards, cryptocurrency, wire transfers, payment apps or cash handed to a courier. See How Scammers Tell You to Pay.

AI can fake a voice, but it can’t change the fact that a real family member, bank or agency doesn’t need you to pay that way in the next ten minutes.

Habits that beat AI-powered scams

1. Verify through a second channel. Hang up and call the person or company on a number you already know. The FTC advises not trusting the voice on the line, even if it sounds exactly right.
2. Agree on a family safe word. The FBI recommends a secret word or phrase to confirm identity in an emergency. Our family emergency scams guide explains how.
3. Treat celebrity investment videos as ads, not advice. No legitimate investment promises guaranteed returns. See crypto investment scams.
4. Limit public audio and video of yourself and your kids, or keep accounts private, so there’s less material to copy.
5. Don’t sign in or pay from a link in a message. Open the app or type the website yourself.
6. Protect your accounts with unique passwords and two-step verification, so a convincing phishing message alone can’t take them over. Helping a parent? See setting up a password manager for an older parent.

Can you spot a deepfake?

Sometimes. Odd blinking, mismatched lip movements, blurry edges around the face, strange hands or a flat, emotionless voice can be clues. But the technology improves quickly, so don’t rely on spotting flaws. Rely on verifying the request instead. Our article on misinformation and deepfakes has tips to share with kids.

FAQ

How do scammers get my family’s voices?

Usually from public videos on social media, voicemail greetings or short phone calls. A few seconds of clear audio can be enough.

Should I stop answering unknown numbers?

Letting unknown calls go to voicemail is a reasonable habit. If it’s important, the caller can leave a message, and you can call back on a number you trust.

Are AI chatbots themselves dangerous?

Legitimate chatbots aren’t scams, but fake “AI assistant” apps and investment bots are common. Download apps only from official stores, and don’t share passwords or financial details with any chatbot.

Where do I report an AI scam?

Report it at ReportFraud.ftc.gov and to the FBI at IC3.gov. If you lost money, call your bank first.

Related reading

• Why Smart People Fall for Scams
• Imposter Scams
• Phishing Text Messages: Examples and Red Flags
• Scam A to Z

Sources

• Federal Bureau of Investigation, Internet Crime Complaint Center, “Criminals Use Generative Artificial Intelligence to Facilitate Financial Fraud,” ic3.gov, accessed November 9, 2026.
• Federal Trade Commission, “Scammers use AI to enhance their family emergency schemes,” consumer.ftc.gov, accessed November 9, 2026.
• Federal Trade Commission, “Fighting back against harmful voice cloning,” consumer.ftc.gov, accessed November 9, 2026.
• Federal Trade Commission, “How To Avoid a Scam,” consumer.ftc.gov, accessed November 9, 2026.

This is general information, not legal or financial advice. If you lost money, contact your bank right away and report the scam.