Elon Musk discusses new company xAI, ‘superintelligence,’ and China

Therefore, he decided to “develop artificial intelligence in a good way”. For Musk, that means training it in a way that is “maximum curiosity, maximum truth-seeking.”

“I think an AI that’s extremely curious, an AI that’s just trying to understand the universe, would be beneficial to humanity,” Musk said. “From the perspective that it’s more interesting to be human than not.”

Musk said he prefers this approach to “trying to explicitly program ethics into AI.” This approach, he said, requires judgment about whose morals are programmed into the system, and also creates an “inversion problem” in which morals could be reversed.

As far as regulatory technology is concerned, Musk suggested that policymakers spend a few years learning about the technology before “monitoring and consulting with industry.” He also suggested that some kind of industry group or self-regulatory body, similar to the Motion Picture Association, would be useful.

The trio’s discussions focused on the threat of China getting ahead of the United States in artificial intelligence and the possibility of some form of military confrontation over Taiwan’s sovereignty. Taiwan is an autonomous region that the Chinese government considers to be part of its territory. territory.

Musk said he spent a lot of time discussing AI safety in conversations with senior Chinese leaders during a recent trip to China. The idea that a “digital superintelligence” could replace the Chinese Communist Party itself seemed to resonate, he said.

“No government wants to find itself replaced by a digital superintelligence,” he said. “So I think they’re actually taking action on the regulatory side and worrying that’s a risk.”

Musk even said he believed the Chinese government was willing to cooperate on an international framework for AI regulation.

Gallagher, chairman of the China Select Committee, pushed back on the notion that the CCP could be a constructive member of this international framework. He warned that even if they heeded Musk’s warnings, he feared it wouldn’t be enough to slow down their efforts on artificial intelligence, which Chinese leader Xi Jinping would use to entrench “totalitarian control.”

After briefly changing the subject, Musk returned to Gallagher’s thoughts, calling himself “pro-China.” Musk acknowledged that he “has some vested interest in China,” but ultimately argued that “China is undervalued” and that “the Chinese people are really great.” Gallagher later said he was “all for the Chinese people” but he was against the ruling party.

While tensions between the U.S. and China remain high, the two countries have made some diplomatic progress by resuming face-to-face communications, with Treasury Secretary Janet Yellen’s recent visit to China and U.S. climate envoy John Kerry due later this month. visit China sometime.

Still, Musk predicted that resolving the Taiwan issue would be difficult.

“I do have a theory about predictions that the most interesting outcomes — as seen by a third party rather than the participants — are the most likely,” Musk said. “It does suggest that the Pacific Ocean is likely to get hot. Hopefully not too hot. But it’s going to be hot.”

Musk predicts that the world will see a digital superintelligence, or an artificial intelligence “smarter than any human being at everything,” within five or six years.

“It’s not necessarily smarter than all of humanity put together,” Musk said. “It’s a much higher bar to be smarter than a few humans. Especially considering it’s the sum total of all machine-augmented humans, because we’re going to have computers, phones, and software applications. We’re already de facto cyborgs.” Humans, it’s just that computers haven’t integrated with us yet.”

“If I were to assign probabilities, I think it’s more likely to be a positive scenario than a bad one,” he said. “It’s just not 0% bad things, we want to do everything we can to minimize the chance of bad outcomes from AI.”

Musk at one point described the future as a TV series.

“If it was a Netflix series or something, I think the season finale would be a showdown between the West and China, and the show would end with AGI” (artificial general intelligence), he said.