In just a year, artificial intelligence has gone from a sci-fi movie to a tool that helps us improve our resumes and plan European vacations.
Given the rapid development of AI models like OpenAI’s ChatGPT and Google’s newly released Gemini, some may wonder whether these systems could eventually replace humans entirely.
But many tech experts don’t seem too worried about this happening any time soon.
“AI can certainly recognize your pet cat, but it won’t solve world hunger,” Theo Omtzigt, chief technology officer at Lemurian Labs, tells CNBC Make It.
AI is unlikely to replace humans because of mathematics
One reason AI is unlikely to completely replace humans is both fairly simple and complex: mathematics.
Large language models, a subset of generative AI, rely on powerful mathematical formulas to process and identify patterns in massive amounts of data to transform user input into new text, image, video or audio output.
But human intelligence goes far beyond pattern recognition. That’s why the mathematical models that drive current generative AI systems are “relatively super simple,” says Omtzigt.
“Right now the machine is learning how to recognize a cat and what it looks like in different light,” he says. “We would have to advance much deeper in our understanding of creative thinking, ethics and consciousness before we would even have the building blocks to think about how we might create an AI capable of wiping out humanity.”
AI systems acquire knowledge differently than humans
Another reason technology experts don’t believe AI will replace humans is because it acquires knowledge differently than humans.
“Generative AI and machine learning techniques are very much based on correlation rather than causation,” Justin Lewis, vice president of incubation and engineering at BP, said Thursday during a panel discussion at the AI Summit New York 2023.
After processing many rain images, an AI model may learn to correlate rain with clouds because clouds are present in every rain image. However, a human learns that clouds make rain, says James Brusseau, a philosophy professor at Pace University who also teaches AI ethics at the University of Trento in Italy.
“AI and humans are both knowledge producers, just as the sculptor and the painter are both artists,” he tells CNBC Make It. “But in my opinion they will forever be different and separate. One will never be better than the other, they will just be different.”
AI will not replace humans, but humans who can use it will
Fears that AI could replace humans are not entirely unfounded, but it will not be systems alone that take over.
“There will definitely be a difference between those who use AI and those who don’t,” said Trevor Back, chief product officer at Speechmatics, during Thursday’s panel discussion at AI Summit New York 2023.
“If you don’t use AI, you’re going to struggle because most roles use some form of AI in the way they trade,” he said.
For example, many tech experts currently see the use of AI as a tool to help people increase their productivity. A software developer can use AI to speed up the code review process and identify potential errors that he or another human may have missed.
Since AI doesn’t appear to be going away any time soon, one of the best ways to ward off the fear of being replaced by AI is to figure out how these systems work and how they can help you, says Brusseau.
“Curiosity is good,” he says. “You have to say, ‘I’m not afraid of what the machine might do to me. I’m interested in what she can do for me.’”
DON’T MISS: Want to be smarter and more successful with your money, your work and your life? Sign up for our new newsletter!
Receive CNBC’s free Warren Buffett guide to investingwhich brings together the billionaire’s best advice for regular investors, rules of conduct and three key investing principles in one clear and simple guide.
CHECKOUT: As ChatGPT celebrates its one-year anniversary, it has more than 1.7 billion users – here’s what it might do next
Source : www.cnbc.com