Make AI Machines Safe Before Smart, Experts Say
Researchers need to develop measures to guarantee the safety of artificial intelligence before businesses and consumers widely adopt it, experts said on Friday.
"It's crucial that we solve the control problem before we figure out the solution to the problem of how to make machines really intelligent," Nick Bostrom, director of the Future of Humanity Institute at Oxford University, told CNBC's "Closing Bell."
Microsoft co-founder Bill Gates became the latest prominent business leader to warn on the dangers of artificial intelligence this week, saying that AI could eventually grow "strong enough to be a concern." Tesla Motors CEO Elon Musk has repeatedly sounded the alarm on the technology, as well.
People first need to realize that widespread adoption of "machines that are intelligent as humans" is not imminent, Bostrom said. Developers need to take care of safety and control measures before that happens, he argued.
The machines can potentially provide invaluable economic aid, said Max Tegmark, an associate professor of physics at the Massachusetts Institute of Technology. Humans should not prevent the progression but rather focus on reducing the concerns expressed by Gates, Musk and others.
"Before we create something smarter than us we need to do the research to make sure we do it carefully," Tegmark said.
He added that technologies such as self-driving cars have already started humans down the path toward AI.