Can You Spot AI-Generated Emails? Why It’s Important & How to Do It

Even though there is a lot of talk about the rapid advancements in AI technology, most people have not yet tried out these Large Language Models (LLMs) for themselves. 

In 2023, after the public introduction of ChatGPT, there was a huge surge in the number of people who had used the site. Even so, the vast majority of Americans have never used it. For people who stay up to date with AI technology and use it for work, ChatGPT and other platforms are an everyday part of life, but most Americans have little interaction with these tools. 

This is probably why so many people struggle to identify AI-generated content when they encounter it. 

Being able to spot AI-generated emails, chatbots, social media posts, and other content can help you avoid falling for scams and hoaxes. Not all AI-generated content is bad, of course–but everyone would be safer online if they knew whether they were interacting with AI bots or real people. 

Why is it difficult to spot AI-generated writing?

The capabilities of these LLMs are improving all the time. With just a few prompts from the user, they can produce huge amounts of writing that mimic human writing very closely. 

These platforms can produce different tones, voices, and levels of personalization, which opens up a world of possibilities for businesses, organizations, and marketers to create more content with less human effort. They can leverage AI to streamline their customer service process, create new marketing campaigns, and personalize their automated communication channels.

Unfortunately, these same strategies also give an advantage to unscrupulous people who want to make some quick money at the expense of their victims! 

As AI becomes more prevalent, it is increasingly difficult for email recipients to distinguish these chatbot-authored messages from those written by actual humans. 

How scammers use AI

Scammers are smart, and they are always learning new ways to trick people into giving them money, personal information, or other sensitive data. 

Let’s take a look at some of the ways that they can use AI to accomplish these goals. 

Highly personalized phishing emails

Phishing emails used to be easier to recognize because of misspellings and grammatical errors. Today, AI fixes a lot of those telltale signs of a scam. 

Scammers also create highly personalized phishing emails, using personal details and contextual information to make the message appear authentic. If you interact with a chatbot that the scammer programmed, you may not realize you are talking with an AI platform as opposed to a real person. 

Realistic tone and writing style 

Because AI is effective at mimicking the nuances of human writing, scammers can use this to create a “personal” that is most likely to persuade you to do what the scammer wants. 

For example, when a scammer needs to pretend to be someone you know personally, like an old friend, they can program their AI output to be friendly and conversational. If they want to pretend to be your boss, the voice can be serious and professional. And if they want to scare you by impersonating an authority figure such as someone in law enforcement, they can use a serious and threatening tone. 

AI helps them to create a more effective scam. 

Spoofing the sender’s email address

Even though AI content is new, the practice of spoofing an email account to make the email look like it was sent by a trusted contact is not. When scammers combine AI-generated emails with spoofed accounts, they can trick a lot of people at one time. 

Drastically increasing how many scams they can perform at once

One of the biggest benefits AI offers to the perpetrators of online scams is scale. They no longer need to respond personally to people who respond to their phishing emails. Instead, the AI chatbot responds. This means they can target thousands of people in the time it used to target a hundred. 

Even though they still only trick a small percentage of people who receive their scammy emails, that small percentage includes a lot more people now. 

How to spot AI-generated emails 

So how do you protect yourself from becoming a victim of one of these scams? You can learn to detect AI-generated emails!

Here are 10 tips for spotting AI:

• Inconsistent tone and style. Real people have a unique style and voice. AI-generated content often sounds overly formal, robotic, or inconsistent in its tone.
  
• Lack of personalization. Legitimate emails from a company or an individual will typically address you by name. Sometimes generic greetings like “Dear customer” or “Hi, there!” can be a sign of an AI-generated message.
  
• Unusual sense of urgency or lots of emotion. Scammers often try to create a sense of fear, urgency, or excitement when they contact you. That’s because they want to pressure you into clicking a link, responding, or sharing sensitive information with them. Be wary of emails that are especially emotional.
  
• Questionable links and attachments. Whenever you receive an email with a link, hover over the link to see where it goes. Don’t open attachments unless you are absolutely sure that they are safe.
  
• Inconsistent branding. If the email claims to be from a company you are familiar with, compare the branding to communications from them that you know are legitimate. If you spot differences in the design, logos, and overall branding, you are probably dealing with a scam. AI doesn’t do a good job of copying branding. 

Developing the ability to spot AI-generated emails is an important skill! It will become more important as these technologies continue to evolve. By learning how to spot some of the telltale signs of AI, you can protect yourself from the growing risks of AI-powered scams.