Elon Musk says there’s a ten% to twenty% probability that AI ‘goes bad,’ even whereas he raises billions for his personal startup xAI

During an interview at Saudi Arabia’s Future Investment Initiative summit in Riyadh, Musk sounded the alarm on the risks of artificial intelligence.

“It’s most likely going to be great,” he said. “There’s this sub chance, that could be 10% to 20%, that it goes bad. The chances aren’t zero that it goes bad.”

The risk of a rogue AI could be heightened if the technology grows exponentially as Musk expects it to. He said during the summit that AI is going to get 10 times better this year, and, in another of his lofty forecasts, predicted that AI would be able to do anything that a human can do within the next year or two. 

“How much longer than that does it take to do what all humans can do combined? Not long, probably three years from that point,” he added.

Musk’s forecast about the size and danger of AI comes as his company xAI is reportedly in talks to raise a new funding round that would value it at about $40 billion, the Wall Street Journal reported, citing people familiar with the matter. The company, which created the Grok AI chatbot that X premium members receive if they pay for a subscription, already raised $6 billion earlier this year at a $24 billion valuation. 

Musk famously helped launch ChatGPT maker OpenAI in 2015, but left just three years later over differences in opinion with the other cofounders—and because he said he wanted to build a competing AI company within Tesla, according to OpenAI. OpenAI is now one of the most valuable private companies in the world after it raised $6.6 billion at a $157 billion valuation earlier this year.

Among his other predictions, Musk also said during the summit that by 2040, humanoid robots will outnumber people on earth, and that every country will have “AIs or multiple AIs.” Tesla revealed the latest version of its Optimus bot earlier this month, although it is still not fully run by AI and requires humans to operate it.

Musk, by his own account, is “pathologically optimistic,” and in the past has shrugged off questions about not meeting the ambitious timelines he sets for his EV company Tesla.

Still, he hasn’t been shy in the past about pointing out AI’s potential risks. In May, Musk responded to a Breitbart article on X quoting Nobel Prize winner Geoffrey Hinton’s warnings about the dangers of AI. And he reiterated his warning about AI during the summit this week. 

“I think AI is a significant existential threat and something we should be paying close attention to,” he said.