Billionaire entrepreneur Elon Musk has issued a stark warning that artificial intelligence (AI) has the potential to destroy civilization, citing the need for a new “pro-human” AI option that seeks the truth.
In a recent wide-ranging interview on Fox News’ “Tucker Carlson Tonight” that aired Monday night, Musk cautioned that existing AI technologies, such as Microsoft-aligned “ChatGPT” and Google-owned “DeepMind,” are being trained to be politically correct, which he believes is leading to an AI dystopia.
Musk noted that AI has the potential to destroy civilization and called for a new AI option that seeks truth. His concerns stem from the singularity, a point at which AI surpasses human intelligence and becomes impossible to predict.
“What happens when something vastly smarter than the smartest person comes along in silicon form? It’s very difficult to predict what will happen in that circumstance,” he said. “It’s called the singularity. A singularity, like a black hole, because you don’t know what happens after that.”
The singularity is a possibility, according to DeepMind CEO Demis Hassabis, who recently noted in a “60 Minutes” interview on CBS that AI could reach a state of “self-awareness.” His comments are noteworthy given that last year, Google fired an AI researcher who claimed its LaMDA language model AI had gained sentience. Google said the claims were unfounded.
Musk emphasized the need for caution with AI and called for government oversight, citing AI as a “danger to the public.”
“AI is perhaps more dangerous than, say, mismanaged aircraft design, or production maintenance, or bad car production, in the sense that it is, it has the potential—however small one may regard that probability, but it is nontrivial—it has the potential of civilizational destruction.”
If regulations aren’t put in place until “something terrible has happened,” then it “may be too late.”
“The AI may be in control at that point,” he added, noting that it is “absolutely” possible for AI to take control and make decisions for people. “That’s definitely where things are headed for sure,” he said.
Musk Says Google Co-Founder Seeks ‘Digital God’
Musk also discussed his conversations with Google co-founder Larry Page about AI safety. Musk said he expressed concern that Page wasn’t taking the risks seriously and noted that Page wanted to usher in an AI superintelligence or a “digital god” as soon as possible.
“He really seemed to want digital superintelligence, basically digital god, as soon as possible,” Musk said.
“He’s made many public statements over the years that the whole goal of Google is what’s called AGI, artificial general intelligence, or artificial superintelligence,” he added. “I agree with him that there’s great potential for good, but there’s also potential for bad.
Musk said he advocated for protecting humanity, to which Page allegedly called him a “specist.”
“I wasn’t the only one there when he called me a specist. Yes, I’m a specist, okay. You got me,” Musk quipped. “But what are you?”
“That was the last straw. At the time, Google had acquired Deep Mind, so Google and Deep Mind had about three-quarters of all the AI talent in the world,” Musk added. “They obviously had a tremendous amount of money and more computers than anyone else. So I’m like, we’re in a unipolar world here.”
Need for ‘Truth-Seeking’ AI
Musk initially co-founded OpenAI to “serve as a counterweight to Google” but left the company in 2018 after a power struggle. Since then, the open source organization has become a for-profit company “closely allied with Microsoft,” to Musk’s discontent.
“In effect, Microsoft has a very strong say, if not directly controls OpenAI at this point,” he said. “So, you really have an OpenAI–Microsoft situation, and then Google–Deep Mind are the other two sort of heavyweights in this arena.”
The tech billionaire, who has often expressed his concern about the risk AI poses for the future of human civilization, spoke of his ambition for a third alternative in the AI landscape.
Musk said that while he was “late to the game,” he will try to create a “maximum truth-seeking AI trained to care about humanity and be ideologically neutral and seek to understand the universe’s nature.
“We want pro-human. Make the future good for the humans. Because we’re humans,” Musk said.
In his view, an AI that cares about understanding the universe would be less likely to destroy civilization because humans are an intriguing part of the universe. He cited the example of humanity’s choice not to exterminate chimpanzees as an illustration of why an AI with such an interest would not view humanity as a threat.
“The intention with OpenAI was to do good. But it’s not clear whether it’s actually doing good or whether it’s … I can’t tell at this point,” he said. “I am worried about the fact that it’s being trained to be politically correct, which is simply another way of being untruthful, saying untruthful things.”
Training AI to be politically correct and therefore “deceptive” is a “path to AI dystopia,” he added.
“So, I’m going to start something which I call ‘Truth GPT’ or a maximum truth-seeking AI that tries to understand the nature of the universe,” he said. “I think this might be the best path to safety, in the sense that an AI that cares about understanding the universe is unlikely to annihilate humans because we are an interesting part of the universe.”