Defining Artificial Intelligence
As we enter what some may call “The Golden Years” of Artificial Intelligence, it’s important to understand just how much AI can do. Most people know AI by its common forms such as ChatGPT, Copilot, Grammarly or resume/cover letter builders. Programs such as Grammarly and websites that assist in building a resume or cover letter are known as “assistive AI.” ChatGPT and Copilot can also be used in an assistive way but are mainly used to generate something based off what the user asks for. When AI is generating something, that is called “generative AI”.
As generative AI has risen in popularity, so have security risks. While the most common uses of AI are for simple tasks such as creating a quick illustration or professional email, that does not negate the fact that people use it maliciously. As generative AI has evolved, it is possible for the human voice to be replicated. Again, common uses for this could be podcasts, films, audiobooks, game production and even music. However, because of this evolution, the likelihood of malicious use is never 0%.
Voice-Cloning Scams with AI on The Rise
For example, there was a security breach regarding AI replicating the human voice via phone call, in 2019. Phone scams are typically nothing new, but the target is rarely an accomplished CEO. A UK-based CEO of an energy firm believed that he was discussing financial matters with the chief executive of the energy company’s parent company that is based in Germany. He was requested to transfer a total of $243,000 to a Hungarian supplier’s bank account—which he did without questioning. According to the CEO, he recognized the subtle German accent and the same cadence, hence why there was no questioning on his end. Unfortunately, it is not known whether this scammer was ever caught.
A more recent instance of a scam that used AI to replicate a human voice was in 2023. This scam targeted senior citizens in which a grandchild or other loved one would contact the grandparent asking for immediate financial assistance. This type of scam conned senior citizens out of roughly $3.4 billion, according to the FBI. Unfortunately, the “grandparent scam” is not a new tactic that is used, but the numbers in cases have increased.
How To Protect Yourself and Your Business from Attacks
So, how can civilians and CEOs alike prevent from falling victim to these scams? There are a few ways to do so:
1. Develop a code word/phrase.
By developing a code word or phrase between yourself and the other party, this can immediately expose a scammer who is using AI to clone a voice. It’s essential to not share the code word/phrase with anyone else outside the circle of trusted people. Make sure that this code word/phrase is also something that isn’t commonly said in conversations. So, a street name, alma mater or other information that can be easily found online is not advised. To properly sus out a scammer, you could ask a question that would potentially lead to the other person on the line saying the code word or phrase if they know it. It’s highly important for you not to say the code word or phrase first.
2. Perform or Attend Phishing/Vishing Workshops Regularly
Stay ahead of the curve by performing or attending workshops dedicated to phishing and vishing scams, regularly. Tactics from scammers are ever evolving, so why shouldn’t you? Workshops that cater to these scams can provide interesting insight in an easily digestible way, rather than over-burdening yourself with endless articles.
3. Avoid Oversharing on Social Media
Social media is where we share our thoughts and connect with friends and family. However, it is important to be cautious as to how much you are sharing online. Instead of sharing exactly what you had for a meal at an exact time and place, maybe give a vaguer description of your activities. The less that potential scammers know, the better it will be for your security.
Conclusion
As the CEO of an IT company, your technological security is my number one priority. To me, it’s important that people understand not only the benefits but the potential risks of artificial intelligence. Incorporating the three tips mentioned above into your daily life and business practices are a great start to being able to identify reality from a scam that uses AI.
This story was contributed by Robert Hessel, chief executive officer of Source 1 Solutions, based in Clearwater.