My daughter called me out of the blue and said she needed money while traveling abroad. She added that more details would be provided later. As a parent, the only thing you can do is give up.
In this hypothetical scenario, my money was stolen by a fraudster posing as my child using a digital voice clone created with generative AI. You may think: I know what my child’s voice sounds like. ”But I assure you, if you were on the phone with one of these digital twins, you wouldn’t be able to tell the difference. This spoofing technology has become so advanced.
This is just one example of the challenges consumers, businesses, and governments are currently facing as genetic AI rapidly becomes mainstream. New threats are being created in the cybersecurity field, but so are opportunities. On the other hand, Gen AI has made it much cheaper and easier to create and execute complex forms of fraud. Meanwhile, companies including Mastercard are using Gen AI to add more contextual understanding to their cybersecurity tools, making them smarter, improving the consumer experience, and fighting fraud faster than before. I am.
How to avoid digital twins
A digital twin is a copy of your voice, or a combination of your voice and your video. In either case, the scammer can train the twins to say whatever they want.
Creating these digital clones used to be very expensive. Today, you can sign up for any number of AI-powered services online, upload a series of audio recordings or videos, and create a digital twin virtually free of charge.
In other cases, scammers can use gen AI’s text generation capabilities to create targeted, personalized messages at scale that can reach many victims through messaging apps.
In these examples, fraudsters now have powerful tools at little cost. These scams are already happening. In a high-profile incident in February, a Hong Kong financial employee was tricked into transferring $25 million to fraudsters after conversing via video conference with the digital twins of several colleagues, including the company’s chief financial officer. I was made to do it.
The serious implication here is that this is an unfair battle for us humans. No matter how hard we try, we cannot learn or adapt as quickly as AI.
Returning to the daughter scenario, there is an easy way to deal with this deception. There is a family password that you can use in cases like this. The digital twin does not know that information. Even without a shared password, you can ask a potential digital twin to confirm information that it doesn’t even know, such as restaurants you’ve recently been to.
No matter how smart your digital twin is, no matter how quickly you evolve your practices, suddenly the entire digital twin facade begins to crumble. This is how you avoid trying to run the same race as the AI. Instead, focus on being more human.
In some cases, less high-tech solutions may work just fine to combat these scams.
How to use Gen AI to strengthen your cybersecurity
Of course, Gen AI isn’t just being used by scammers. And low-tech solutions aren’t always the answer. Mastercard teams are now using Gen AI and AI tools to improve security in the digital world.
In another example, more sophisticated fraud models can be built with the help of AI. Currently, many fraud models are built using users’ purchasing habits and behaviors. So if you typically spend $100 a day around Philadelphia, suddenly trying to buy a $2,000 item in the Philippines could flag a fraudulent model.
AI models can be used to understand spending habits more broadly than just at an individual level. Let’s say you’ve never gambled before, but your bank’s fraud system reports gambling expenses. Looking at broader data, AI systems can help identify that many people staying at a particular casino resort are gambling. This gambling expense will no longer be considered suspicious since the transaction shows you are staying at that resort. No more sending alerts or suspending cards. That creates a better experience.
In just a few years, the AI generation will change how we interact with technology, how we write code, and whether what we hear and see is really real, and even when my daughter calls me. Even I started to doubt it.
The story about genetic AI is complex, and the risks are clear. But I know that we can guide it in the right way and that people and businesses can benefit from the new wave of innovation it will generate.
Banner image via Adobe Stock. Created using generative AI.
