Elon Musk’s xAI Will Use Public Tweets for AI Model Training, to Work With Twitter and Tesla

Elon Musk on Friday said the purpose of his new artificial intelligence company, xAI, will be to “understand the universe.”

In an often meandering 90-minute long Twitter Spaces audio chat, the billionaire discussed his vision for xAI for the first time while veering into topics such as the Earth’s evolution and the fragility of civilisation.

In seeking to deepen the understanding of the universe, Musk joked that xAI’s mission statement would be “what the hell is really going on?”

Musk announced the formation of xAI on Wednesday after accusing companies like OpenAI and Google of developing the technology without considering risks to humans.

OpenAI and Google did not respond to request for comment.

He said xAI would seek to build a “good AGI,” as an alternative to Microsoft, Google and OpenAI. AGI stands for artificial general intelligence and refers to AI that can solve problems like a human.

During the Spaces session, which had a delayed start as Twitter needed to “tweak the algorithm” to promote the chat to more users, Musk said xAI would work closely with his other companies, Twitter and Tesla.

The company will use public tweets to train its AI models and may also work with Tesla on AI software.

Such a relationship would have “mutual benefit” and could accelerate Tesla’s work in self-driving capabilities, Musk said.

He also accused all AI companies of training their models using Twitter data in what he characterised as an illegal manner.

Musk, who has advocated for regulations in AI, said he has pushed for meetings with White House officials and has emphasised the importance of regulating AI in his recent meetings with top government officials in China.

© Thomson Reuters 2023