But Microsoft executives have had reservations about sending additional funding to OpenAI since 2018 when it was just a small nonprofit research lab, according to emails between more than a dozen Microsoft executives, including CEO Satya Nadella, that appeared in federal court Thursday during the hearing. Musk v. Altman trial.
The emails show how Microsoft hesitated, at the time, about what has since been considered one of the most successful enterprise partnerships in technology history. Several Microsoft executives said in emails that their visits to OpenAI did not indicate any imminent breakthroughs in the development of artificial general intelligence. In 2017, much of OpenAI’s work focused on building AI systems that can play video games, which has shown early signs of success. But OpenAI needed five times more computing power than it had originally gotten from Microsoft to continue the project.
Microsoft worried that not offering support might push OpenAI into the arms of Amazon, the world’s dominant cloud computing provider at the time. Nearly 18 months after the emails were sent, Microsoft announced a massive $1 billion investment in OpenAI after the lab created a for-profit arm that provided the tech giant with the potential to generate $20 billion in revenue.
Microsoft declined to comment.
Elon Musk’s lawyers provided the emails to show Microsoft’s evolving relationship with OpenAI. After Musk approached Nadella, Microsoft agreed in 2016 to provide $60 million worth of cloud computing services to OpenAI at a deep discount. OpenAI consumed services twice as quickly as expected.
The email chain began on August 11, 2017, with Nadella reaching out to Sam Altman, CEO of OpenAI, to congratulate the lab on winning a video game competition using artificial intelligence to mimic a human player. Ten days later, Altman responded by acquiring Microsoft Azure cloud computing services for $300 million.
“We can figure out how to fund some of these projects but not a lot,” Altman wrote, apparently seeking financial aid and engineering assistance. “I think it will be the most impressive thing yet in the history of artificial intelligence.”
Nadella asked four aides for input on how to respond three days later. Microsoft’s AI team saw “no value in participating,” according to a response by Jason Zander, Microsoft’s executive vice president, who also documented how other teams felt. Its research team thought its work was “more advanced,” while its PR teams didn’t like the idea of supporting a group promoting the idea of “machines beating humans.” Ultimately, Zander suggested that Azure would benefit from linking up with Musk and Altman, but he didn’t want to “take a full bath” or take a big financial hit by doing so.
A later analysis showed that Microsoft would lose about $150 million over several years if it provided the services Altman wanted, according to one email. “Unless he can help us draw a more direct network effect with OpenAI -> Microsoft business value, we will end up succeeding,” Zander wrote.
The topic disappeared for several months, but was revived on January 10, 2018, with an email to Nadella from Brett Tanzer – who signed his emails with “Brett” – who was then director of Azure’s cloud unit. Altman told Tanzer that OpenAI could license gaming AI to Microsoft’s Xbox video game division for “$35-50 million in Azure credit.” But Xbox couldn’t allocate that kind of money. Microsoft planned to tell Altman that there would be no more discounts after March, according to Tanzer’s email.
