Elon Musk’s announcement Wednesday that he’s joining the AI race with his new company, xAI, which will set out to “understand the true nature of the universe,” comes after fits and starts with the world’s richest man’s involvement with artificial intelligence. He was an original board member of OpenAI, departed in 2018, and was an an early investor in Deepmind, but before now he has never embraced the A.I. space as a solo pursuit, while often voicing concerns for the dangers posed by AI.
This year, as he’s admitted to overpaying for Twitter by approximately $20 billion, he’s seen his thunder in Silicon Valley stolen, first by OpenAI’s ChatGPT, which became the talk of the Bay and Wall Street as it soared to the fastest adoption of any technological app in history. More recently, he’s seen his rival Mark Zuckerberg launch a “Twitter killer,” Threads, via Meta’s Instagram app, and rack up an even faster adoption rate: Over 100 million users in less than a week.
Musk has acknowledged his chagrin. “I fully admit to being a huge idiot here,” Musk said in a May interview with CNBC about leaving OpenAI before it took off.
It’s a far cry from the last decade, when Musk began calling A.I. one of humanity’s greatest threats. Musk has never held back on providing colorful imagery to explain the potential for an AI apocalypse. In a 2014 speech at the Massachusetts Institute of Technology, he used colorful imagery: “With artificial intelligence, we are summoning the demon. In all those stories where there’s the guy with the pentagram and the holy water, it’s like – yeah, he’s sure he can control the demon.” Musk added that it never works out in those stories.
He also called A.I. and robots powered by AI a “fundamental risk” to humankind at a National Governors Association 2017 Summer Meeting. At the meeting, he warned of AI robots learning too quickly and becoming uncontrollable, as well as spreading misinformation and sharing.
“AI is a fundamental risk to the existence of human civilization in a way that car accidents, airplane crashes, faulty drugs or bad food were not — they were harmful to a set of individuals within society, of course, but they were not harmful to society as a whole,” said at the conference. He also called for regulation and cohesive oversight of the technology.
In 2017, Musk tweeted that the race for AI superiority would cause such strong competition between nations that it could cause World War III. The tweet came in response to an article in The Verge, which said that Russian President Vladimir Putin had predicted that world supremacy would fall to the country that created the first A.I. That same year, Musk said A.I. is more extreme a danger than North Korea testing a hydrogen bomb.
China, Russia, soon all countries w strong computer science. Competition for AI superiority at national level most likely cause of WW3 imo.
— Elon Musk (@elonmusk) September 4, 2017
If you’re not concerned about AI safety, you should be. Vastly more risk than North Korea. pic.twitter.com/2z0tiid0lc
— Elon Musk (@elonmusk) August 12, 2017
As the A.I. space became the hottest trend in technology, Musk called for a pause in development, citing concerns and calling for regulations.
His concerns are hinted at on xAI’s website, which says it is going to be advised by Dan Hendrycks, the director of the Center for AI Safety, which has warned of the dangers of A.I. technology that develops too rapidly. But Musk also has recruited several members of the xAI team from rival A.I. firms, including Google’s DeepMind and Microsoft.
He further specified his thoughts on the competition in the May CNBC interview, pointing to a fallout with Google’s Larry Page about Microsoft’s ownership and Page being “cavalier” with regard to the technology. He added in the interview he also disagrees with companies profiting from A.I.
“I do worry that Microsoft actually may be more in control than, say, the leadership team at OpenAI realizes,” he said in the interview, citing concerns for profit-driven A.I. technology at Microsoft.
Musk also unveiled a robot named Optimus at Tesla AI Day last year, that would use similar A.I.-assisted navigational features to the Tesla. He also spoke ominously at a Tesla conference in March, saying once AI can help build Tesla cars, there’s no point in anyone working.
“I don’t know. Tesla is doing good things in AI,” he said at the conference, according to Reuters. “This one stresses me out. I don’t know what to say about it.”
Musk’s full outlook on running an AI company and his intent with AI is certainly to come. Right now, xAI’s website doesn’t say much more than list people and a goal, but it does show that Musk isn’t just sitting on the sidelines watching the AI brawl play out between the other tech giants.