An AI scam is a fraud scheme that uses artificial intelligence to deceive victims by creating convincing, machine-made content. While similar technology has existed before, little compares to the ease of access, volume, and scale of content AI can churn out. These sophisticated scams often fall into one or more of these categories: text, voice, and video. 

• Smishing: Text-based scams in which criminals send hyper-personalized messages, often via text or email.

• Vishing: Voice-based phishing scams that clone a trusted individual’s voice in an audio message or phone call, either pre-recorded or live.

• Deepfakes: Generated or altered images and videos that convincingly depict family, friends, or public figures. Modern deepfakes can replicate details with high accuracy, often making them indistinguishable from reality.

What Do AI Scams Look Like in Practice?

AI scams can take many forms, depending on whether they’re text-, voice-, or video-based. Some common examples include: 

• Distress Scams: Scammers use voice cloning to mimic a loved one, family member, or close friend in distress. These scams use AI to react in real time to your responses and rely on emotional manipulation, such as telling you not to hang up, invoking fear, or pressuring you with urgency.

• Unexpected Requests: Criminals imitating a trusted individual unexpectedly ask you for money, gift cards, passwords, personal information, or to join an investment opportunity.

• Business Email Compromise (BEC): Scammers impersonate trusted leaders to trick employees into sending money or data. BEC scams may occur through virtual interviews and meetings, with the primary goal of accessing private computer networks.

• Romance Scams: Through text or social media, scammers build a friendly rapport over time and eventually ask for money, preying on people's friendliness. 

How to Protect Yourself 

Follow these tips to know what to do if you receive a suspicious message and how to prevent future AI scams. 

1. Take a Beat: Scammers want you to make quick, impulsive decisions, which is why distress scams are common. Instead, take a beat to assess the situation before responding to suspicious messages.  

2. Look for Inconsistencies in Deepfakes: Always pay attention to details in suspicious videos to spot uncharacteristic communication and errors. For example, does the person blink too much or too little? Is the voice tone flat or unnatural? Do the visuals show strange or unnatural shadows or lighting?  

3. Verify Sources: Before sending money or sharing personal information, always verify the source. Contact the person being deepfaked another way, such as using a phone number you’ve previously used. Be careful, as scammers may attempt to control how you contact others by claiming their old phone is unavailable. Also, consider establishing a group secret codeword for real emergencies.  

4. Multi-factor Authentication: If scammers get hold of your information, strong security can delay or prevent further damage. Always use strong multi-factor authentication and unique, complex passwords for each website.  

5. Don’t Feed the Data Monster: Generative AI models are built on datasets, and scammers only need a few seconds of your audio, a few photos, or a few seconds of video to create deepfakes. Limit your digital footprint and never answer a scam voice or video call.

While technological advancements have made generative AI content easier to create and nearly indistinguishable from reality, staying informed and prepared is key to protecting you and your loved ones from criminal actors. 

At Southern First, your safety is our top priority. If you receive a suspicious video, call, text, or email, please contact our Fraud Department at 1-877-679-9646. We are here to help.