AI Scams: 8 Common Types and How to Spot Each One

AI scams are frauds that use artificial intelligence to make deception cheaper, faster and more convincing. The most common types are AI-written phishing, cloned-voice phone calls, deepfake video, fake chatbot support agents, romance and friendship scams run with AI help, fake job and scholarship offers, AI-generated investment pitches and fake reviews or shops. The underlying tricks are old; AI just removes the spelling mistakes and makes impersonation easier.

This guide explains how each scam type works, then gives you a four-step routine to check whether a specific message, call or video is real. The short version: judge the request, not the polish, and verify through a channel you already trust.

Why AI changes the scam landscape

Scams have always relied on three things: a believable story, a sense of urgency and a way to get paid. AI helps with the first two.

• Polish at scale. Generative text tools write fluent messages in any language. The clumsy grammar that once gave scams away is no longer a reliable clue.
• Personalization. Scammers can feed public information, like your job title or recent posts, into a tool and get a message tailored to you.
• Impersonation. Voice cloning and video deepfakes let criminals sound or look like someone you trust.
• Conversation. Chatbots can keep up long, patient conversations with many victims at once.

What AI does not change is the goal. Nearly every scam ends with a request for money, gift cards, cryptocurrency, passwords, verification codes or personal documents.

1. AI-written phishing emails and texts

Phishing messages pretend to come from a bank, delivery company, university, employer or popular service. They push you to click a link, open an attachment or "verify" your account.

With AI, these messages are well written, match the tone of the real organization and may mention details about you. Some arrive as text messages, others in workplace chat apps.

Typical hook: "Unusual sign-in detected. Confirm your identity within 24 hours to avoid suspension."

Warning signs: a link that does not match the official domain, pressure to act fast, requests for login details or codes.

2. Cloned-voice phone calls

A scammer collects a short recording of someone's voice, often from public videos, and uses a voice-cloning tool to imitate it. Then they call a relative or colleague with an emergency: an accident, an arrest, a stolen wallet abroad.

Phone audio is low quality, and panic makes people less critical, so a decent imitation can be enough.

Typical hook: "Mom, it's me. I'm in trouble, please don't tell Dad. I need money for bail right now."

Warning signs: urgency, secrecy, unusual payment methods and an inability to answer a personal question.

3. Deepfake video calls and clips

Deepfake video puts one person's face (and sometimes voice) onto another's. There have been widely reported cases of employees being persuaded on video calls with what appeared to be senior executives to transfer company funds.

Deepfake clips also appear in ads, showing celebrities or public figures endorsing products or investment schemes they never mentioned.

Typical hook: a "CEO" on a short video call asking for a confidential urgent payment, or a famous person "recommending" a trading app.

Warning signs: odd lighting or lip sync, refusal to turn the camera off and on or move, out-of-character requests and pressure to skip normal approval steps.

4. Fake customer support chatbots

Scammers set up sites or social media accounts that look like official support for banks, airlines, marketplaces or software companies. An AI chatbot answers your questions convincingly, then asks for card details, remote access to your device or a "refund verification" payment.

These often appear in search ads or as replies when people complain publicly about a company.

Warning signs: support reached through an ad or a social media reply rather than the company's official site or app, and requests for remote access or payment to receive a refund.

5. Romance and friendship scams

Long-running scams that build trust over weeks or months have become easier to run at scale. AI helps write warm, consistent messages, translate them naturally and generate profile photos of people who do not exist.

Eventually the "partner" or "friend" needs money for an emergency, a visa or a trip, or introduces an investment opportunity.

Warning signs: someone who never meets in person or on a live, natural video call, profile photos that look slightly too perfect, and any request for money or investment.

6. Fake jobs, internships and scholarships

These scams target students, recent graduates and job seekers. AI makes it easy to produce professional-looking job descriptions, offer letters, company websites and even chatbot "interviews."

The catch comes later: a fee for training, equipment or a background check, a request for banking details "for payroll," or a check you are asked to deposit and partly send back.

Typical hook: "Congratulations! Based on your profile, you've been selected for a remote data-entry internship. No interview needed."

Warning signs: offers you never applied for, no real interview with a person, payment required up front and communication only through chat apps.

7. AI-powered investment schemes

Fraudsters promote trading bots, crypto platforms or "AI-driven" funds promising unusually high, steady returns. They use AI-generated testimonials, deepfake endorsements and polished dashboards that show fake gains.

When victims try to withdraw, new fees appear or the platform disappears.

Warning signs: promised returns that sound too good, pressure to recruit friends, unregistered platforms and withdrawals that keep getting delayed.

8. Fake reviews, shops and listings

AI can write hundreds of believable product reviews in minutes, create product photos and build entire online stores. Shoppers pay for items that never arrive, or receive something very different.

Rental scams follow a similar pattern: an attractive apartment, AI-written listing, generated or stolen photos and a "landlord" who cannot meet but needs a deposit.

Warning signs: prices well below market, reviews that all sound the same, a brand-new site with no history, and payment only by bank transfer or gift card.

Scam types at a glance

What all AI scams have in common

Despite the technology, the pattern rarely changes:

1. A trusted identity: a bank, a boss, a family member, a famous person.
2. An emotional trigger: fear, excitement, love, embarrassment or greed.
3. Urgency or secrecy: act now, tell no one.
4. An unusual payment method or request: gift cards, crypto, wire transfers, verification codes, remote access.

How to spot an AI scam: verify the source, not the polish

For years, people were taught to look for bad grammar, strange logos and generic greetings. Those clues still catch lazy scams, but generative AI writes fluent, personalized text in any language, and image tools produce clean logos and documents. A cloned voice or a familiar face on a video call is just as hard to judge under stress and over a weak connection.

So the strategy shifts from inspecting the content to verifying the source. Use this routine for any unexpected message, call or video that asks you to do something.

1. Pause. Scams depend on speed. Tell yourself, or the caller, that you will check and get back to them. A legitimate contact will understand; a scammer will push harder, and that pressure is itself a signal.
2. Identify the request. What exactly does this message want? Be suspicious of any request to pay (especially by gift card, crypto, wire or payment app), share a password or one-time code, install software or allow remote access, send ID or bank documents, or keep the request secret.
3. Verify through a separate channel you find yourself, not one provided in the message. Open the bank's official app or call the number on your card. Message your manager on your usual work channel. Hang up and call your relative's saved number. Log into your school's portal directly.
4. Inspect links, senders and payments. Hover over or long-press links to see the real address and check the domain for extra words or swapped letters. Look at the full sender address, not just the display name. Check whether a payment recipient's name matches the organization it claims to be.

Extra checks for calls and video

• Do not trust caller ID; numbers can be spoofed. Ask a question only the real person would know, and not something visible on social media.
• On a video call, ask the person to do something unplanned, like turning their head fully or waving a hand across their face. Some deepfakes struggle with this, though the tools keep improving, so treat it as a hint, not a test.
• For any financial request on a call, confirm through a separate channel and follow normal approval rules, no matter who appears to be asking.

What AI detection tools can and cannot do

It is natural to wonder whether an AI detector can catch a scam. Usually it cannot. A text detector estimates whether writing resembles AI output, and many legitimate companies use AI to draft emails while many scams are written by people, so "AI-written" and "scam" are not the same thing. Deepfake detectors for images and video exist, but results vary and they are not a substitute for verifying identity. A fact-checking tool can help you test a claim inside a message, such as whether a policy or news event is real, but it cannot confirm who sent it. If you want to understand what text detectors actually measure, see our guide to how AI detectors work.

A worked example

A student receives a text: "Your tuition payment failed. Pay the outstanding balance within 2 hours to keep your course registration," followed by a link. It uses the university's name and the correct semester.

1. Pause. Two hours feels urgent, but a registration problem can be checked in five minutes.
2. Identify the request. It wants a payment through a link.
3. Verify separately. The student logs into the official student portal from a bookmark. There is no outstanding balance and no alert.
4. Inspect. The link points to a domain that adds an extra word to the university's name.

Result: a scam, identified without any special software. The student reports it to the university's IT security team, which can warn others.

Red flags and safe responses

If you think you have been scammed

Act quickly and without embarrassment. Scams target smart, careful people every day.

1. Stop contact with the scammer. Do not send more money to "recover" losses.
2. Call your bank or card provider right away to block payments or cards.
3. Change passwords for any accounts you shared or that use the same password, and turn on two-factor authentication.
4. Save evidence: screenshots, emails, phone numbers, transaction records.
5. Report it to the platform, your bank and your national fraud or consumer protection agency.
6. Warn people around you, especially if the scammer impersonated you or someone you know.

Reduce your exposure before a scam arrives

A few habits make you a harder target for every scam type on this list.

• Limit public voice and video of yourself and family members where you can, or at least be aware that it can be copied.
• Tighten privacy settings on social media so job titles, school names, travel plans and family details are not visible to strangers.
• Use a password manager and two-factor authentication so a single stolen password does less damage.
• Agree on rules with family and coworkers, such as a code word for emergencies and a policy that no payment is approved on a call alone.
• Talk about scams openly, especially with older relatives and younger students. People who have heard of a scam are much more likely to recognize it.

Key takeaways

• AI makes scams more polished and personal, but the core tricks remain the same.
• The main types are phishing, voice cloning, deepfake video, fake support bots, romance scams, fake jobs, investment fraud and fake shops.
• Watch for trusted identities plus urgency plus unusual payment requests.
• To check a specific message, use the routine: pause, identify the request, verify through a separate channel, then inspect links and payments.
• Code words, call-backs and official apps beat any detector. If something already happened, call your bank, change passwords, save evidence and report it.

Scams are one part of a wider set of problems with how AI gets used; our guide to AI ethics covers deepfakes, privacy and misinformation in that larger picture.