Amber Mac, internet of things (IoT) and artificial intelligence (AI) expert, opened the ACFE Fraud Conference Canada today in Toronto with a hypothetical story called “the paper-clip experiment” to show how helpful and simultaneously dangerous AI could be for individuals and organizations.
The experiment goes like this: What if we used artificial intelligence to train an entire factory of robots and machines to build as many paper clips as they can? Is it possible, then, that if the machines run out of materials to produce the paper clips that they will then look outside the factory to keep the process going? Will they look at and destroy our cars, our homes or any other things that have the materials they need to continue to make paper clips?
This scenario highlights two of Mac’s biggest predictions for the future of technology: that we should be hopeful for all of the possibilities and opportunities that await, but also ready to take on the risks these potential utilities pose.
Mac then walked the more than 300 attendees through a day in the life of IoT and AI in 2025. She described a digitally connected day that began with voice assistance technology (imagine a cheerful good morning message from Amazon Echo’s Alexa reminding you of the meetings you have planned for the day) and an automated car ride (a self-driving car allowing you to catch up on sleep or email).
For some reading this, this futuristic day might already look a little like your present. But, according to Mac, once you look closer at these seemingly useful technologies, you begin to see threats emerge. For example, what would it look like if your company installed voice-assistance technology in your office? What kinds of conversations would you be okay with sharing with technology connected to the internet? Would you trust a voice assistant to be your actual assistant? As Mac said, “I’m now not worried they [hackers] are going to get into my computer versus I am now worried they are going to get into all of my connected devices.”
Mac also described a possible scenario of a hacker penetrating autonomous vehicles. She said, if 94% of traffic accidents are caused by human error, then wouldn’t these vehicles help? Or, conversely, can it think on its feet if there is a car that veers in front of it? And, what happens when a hacker takes over the wheel? Or, gets control of the virtual keys? Mac highlighted a recent news story about an Android app that can track, unlock and steal a Tesla without even having the physical keys.
Mac encouraged attendees to take note of what is to come and prepare for exponential change happening right now. For those of you reading this who feel like you are managing these fraud risks and are sufficiently on top of what is to come, Mac had one bit of advice via race car driver Mario Andretti. To quote the notorious, heavy-footed speedster, “If everything seems under control, you’re just not going fast enough.”