But emails released by OpenAI on Tuesday show that the billionaire agrees that OpenAI will become less transparent as the institute moves closer to achieving its goal of AI matching human intelligence. It seems to be showing. In his message, Musk acknowledged that the only way forward for OpenAI is to raise significant amounts of capital as a commercial company. “Raising hundreds of millions is not enough. This immediately requires billions of dollars a year,” Musk wrote in 2018 after parting ways with OpenAI.
Musk did not immediately respond to a request for comment. An OpenAI spokesperson declined to comment beyond the blog post and said the company plans to ask the court to dismiss Musk’s lawsuit.
A partially redacted message shared by OpenAI sheds light on one of the most closely scrutinized founding stories in Silicon Valley history. Industry insiders quickly took a closer look at the company’s early emails, with some attempting to use modern AI tools to unredact the texts.
OpenAI released the popular ChatGPT in late 2022, kickstarting the current AI arms race. Technology leaders are heralding “generative” AI as a technology change on par with the release of the internet and smartphones. This boom has catapulted OpenAI to the top of the tech industry, enabled it to raise hundreds of millions of dollars in funding, and given it leverage over CEO Sam Altman in the debate over how governments should regulate technology. I was able to give
Musk has consistently questioned whether OpenAI is straying from its roots as a non-profit organization aimed at openly sharing AI research, and he said he was reluctant to start the organization. I cooperated.
Musk’s lawsuit asks a judge to order the company not to profit financially from its AI and to share its research with the wider public. Musk himself has founded his own AI lab called X.ai, and Tesla has been investing in AI for years to develop self-driving cars. Tesla is also developing a humanoid robot called Optimus. ” Musk also complained that OpenAI and its largest partner and investor, Microsoft, collected data for his social media company X to train AI models.
Musk is no stranger to legal battles. He has said in the past that he would fight a legal battle if he believed he was right, even if the chances of winning were slim. OpenAI also faces a number of other lawsuits from authors and news publishers alleging that the company used content to train AI without payment or permission.
In addition to the emails, OpenAI suggests that Musk left OpenAI after his bid to take control failed, rather than over an ideological dispute over openness. “Elon had hoped to acquire a majority stake, gain initial control of the board, and become CEO,” the blog post said. “During these discussions, he withheld funding.” According to OpenAI, venture capitalist and LinkedIn co-founder Reid Hoffman “bridged the gap to cover his salary and operations.” ”.
OpenAI said in its response that Musk was one of the people the company recognized would have difficulty pursuing its mission to develop artificial general intelligence without significant funding. In a message shared by ChatGPT’s developers, Musk pushed OpenAI toward a commercial model and called on the company to “start being open” about the science behind its products. In a 2015 article, he argued that the company should say, “We’re starting with a commitment to raise $1 billion…We’ll cover anything that others don’t.”
OpenAI said Musk failed to fulfill all of his promises, ultimately delivering less than $45 million, but continued to assert control of the company, which underwent a restructuring in 2017.
“We were unable to reach a commercial agreement with Elon because we felt it would be inconsistent with our mission for individuals to have complete control over OpenAI,” the company said in its response. “He then suggested integrating his OpenAI into his Tesla instead.”
One email sent in early 2018, around the same time Musk left OpenAI, shows Musk claiming: Still, it’s unlikely you can compete with Google. ”
In another email published by OpenAI, Musk forwarded a message he had received that appeared to be from Demis Hassabis, the director of Google’s DeepMind AI lab, which Hassabis is the foundation. They have been shown to have expressed concern about the dangers of making AI technology “open source”, meaning it is made available to the public for free. . OpenAI later republished the blog post to remove Hassabis’ name. A Google spokesperson did not respond to a request for comment.
In an email thread, OpenAI’s chief scientist Ilya Satskeva told Musk that as AI advances, companies should release less data about how their technology works. Mr. Musk replied, “Yes.”
Faiz Siddiqui contributed to this report
