Billionaire tech entrepreneur Mark Cuban has seen a ton of change since he first got in the technology business in 1982, but he argues that artificial intelligence (AI) is going to “change everything, 180 degrees.”
He warns that if the U.S. allows other countries to take the lead in AI, then it’ll be “SOL,” an acronym that employs profanity to communicate urgency.
“All these things have happened that have changed how we do business, changed how we lived our lives, changed everything, right, the internet. But what we’re going to see with artificial intelligence dwarfs all of that,” Cuban said in an interview with hedge fund manager J. Kyle Bass of Hayman Capital on RealVision Television, a subscription financial video service.
AI is expected to soon bring an increase in productivity, resulting in fewer jobs all while the population continues to grow.
“It’s not a question of how it plays out over 100 years. It’s a question of how plays out over ten years, 20 years, right? Who knows what exactly? But I can tell you, at the beginning, which jobs are going to be displaced,” Cuban said. “I can tell you the real estate is going to be displaced. I’ve talked to major companies that, they’re asking me, ‘Mark, we’re going to have all this extra real estate in all these towns. What are we going to do with it? Do you have any ideas?’”
Cuban sees AI as deflationary as the increased productivity brings prices down.
“So, then the question, when I sit down with these companies, we’re not doing it just to commiserate how bad things are going to be. The question is ‘what do we do?’”
New types of jobs will emerge
Cuban, who added that he’s spent a lot of time teaching himself about machine learning and writing code, explained that it used to be that people would use spreadsheets and postulate different variables and build models. Smart people would think of more variables, resulting in better probabilities.
Machine learning, however, can think of infinitely more variables than people.
“That’s only the beginning, right? Then, we start getting into something called generative artificial intelligence where it starts to think, right?” Cuban said. “And for the most part, now, not all the way through, you still have to label everything. Here are my variables, right? But then when you give it to it, and you say, based off of these variables that are labeled as such, here’s an unlimited number of opportunities, right? And so based off of those, we’re going to create biases and weights for them all and let you come up with some conclusion. Let it give you a variety of conclusions, and then you decide. Going forward, you won’t have to label anything, right?”