AI Scammers Acquire New and Efficient Tools

As the holiday season gets closer, scammers get their hands on new AI tools to scam consumers. Credit: South_agency via iStock.


Across New York, people have been receiving suspicious phone calls, messages and emails made to look real by scammers using artificial intelligence, which has become more powerful in the months leading up to the holiday season.

“We are living in unprecedented times,” said Roma Majumder, McAfee’s Senior Vice President of Product. “Consumers are navigating a world where advanced AI has exponentially increased the volume and sophistication of scams, making real-time threat and scam protection more important than ever.”

McAfee, one of the major companies that provide computer safety, has recently implemented AI in their own programs to counter suspicious links and phishing emails.

Over the past few years, AI has evolved considerably with people using it for different means, including generating images and videos. However, these tools can also be used to create messages that scammers are using to scam people who are not aware.

ChatGPT is the most common tool these scammers use. By simply entering in a few words, the tool is prompted to generate multiple paragraphs of words in mere seconds. Since its initial launch on November 30th, 2022, the program has learned to mimic real human speech, sounding more convincing every single time. After inserting the prompts, scammers copy and paste the generated paragraphs in an email, include a malicious link, and send them out broadly in hopes of finding unsuspecting victims.

These fake emails are on the rise in New York, with college students being a prime example of who scammers are targeting. Using ChatGPT, the scammer could generate a well-written email that could convince a student that it’s real, making them panic, immediately click the malicious link, and send the scammer their hard-earned money, potentially downloading malware to their device.

A spokesperson for OpenAI, ChatGPT’s parent company, made a statement in an ABC report saying “We don’t want ChatGPT to be used for misleading purposes anywhere.” The company has also developed a program called the Classifier Tool, which can assist people in differentiating emails written by AI to those written by real humans.

Spinoffs of ChatGPT have popped up recently, such as WormGPT and FraudGPT, the latter of which is the newest tool for scammers to use to create malicious phishing emails. This is especially concerning given that we are currently in the holiday season, with Christmas being mere days away. According to the FBI, over $73 million was stolen by scammers in the 2022 holiday season.

Pikes Peak State College AI Policy Chair Dennis Natali made a list of the “12 Frauds of Christmas,” that included fake delivery tracking links and charity/gift card scams. With AI making these scam emails easier and faster to create, the sheer number of them could multiply very quickly.

Barracuda Networks is a company that focuses on protecting people from these phishing emails by identifying them as spam, as well as letting the recipient know by highlighting them in their spam folder. In a research report back in December of 2020, Barracuda discovered that spear-phishing attacks had spiked by 150% in the week before Christmas.

“December is the month of giving, but it’s also the time of year when cybercriminals try to take advantage of consumers, quickly turning a joyous season into a living nightmare,” said U.S. Attorney Dena J. King in a warning to consumers. “Whether shopping online for a gift, choosing a charity for a donation, or applying for seasonal work to earn extra money, it is imperative for everyone to stay vigilant against online scams and identity theft schemes. It’s important for all of us to do all we can to protect our hard-earned money and our personal information from financial predators.”

“One of the biggest pieces of advice that we’ve always given over the years for these phishing attacks, or phishing emails and other things are to look for poorly written emails,” said Alex Hammerstone from tech group TrustedSec in an interview with ABC7 News. “You know, a lot of the people who write these are based overseas, and maybe English isn’t their first language.”

New York Senator Kirsten Gillibrand announced that she would be reintroducing the Deter Obnoxious, Nefarious, and Outrageous Telephone Calls Act (DO NOT CALL Act) and pushing for harsher penalties for robo scam callers, who are using AI to make phone messages look and sound like real people.

“Robocalls are obnoxious and disruptive at best and mechanisms for insidious scams at worst,” said Gillibrand. “Scammers should not be allowed unfettered access to New Yorkers at all times of day and night,” “I am determined to end relentless robocalls and look forward to getting this legislation passed.”