The future of OpenAI is up in the air after the board ousted co-founder and CEO Sam Altman, rattling major investor and partner Microsoft, and potentially creating an opening for rivals.
OpenAI’s board didn’t disclose its specific reason for pushing out the public face of the company on Friday except to say it had lost confidence in Altman’s ability to lead, a day after he appeared at a high-profile event at the Asia-Pacific Economic Cooperation conference.
But internal fighting over the direction of the company’s nonprofit versus for-profit statuses, along with Altman’s influence in comparison to others like chief scientist Ilya Sutskever, who is part of a group at the firm that believes AI can pose a risk to humanity, contributed to the drama, journalist Kara Swisher tweeted.
OpenAI President Greg Brockman, who quit on Friday, said Sutskever informed Altman he was being fired on Thursday night and the company’s management team was informed shortly after the announcement about the CEO’s department was posted.
“Sam and I are shocked and saddened by what the board did today,” Brockman, who quit on Friday, tweeted. “We too are still trying to figure out exactly what happened.”
Semafor also reported that Altman was in the middle of raising money for a new venture capital fund focusing on hard tech when he stepped down as CEO.
He was also in good spirits in recent days and attended an industry gathering on Wednesday evening, other people familiar with the matter said. And he met with new employees this week, giving no signal of turmoil on the horizon, people familiar with the matter said.
OpenAI is no ordinary startup. It was a nonprofit before Altman created a for-profit arm to bring on outside investment from Microsoft. It’s that funding that allowed the company to scale, building the massive large language model that ultimately became ChatGPT and the image model DALL-E.
As Semafor reported in March, Altman doesn’t own equity in the company and some venture capital firms were hesitant to invest because of that fact. Now you know why. Most founders have more power to avoid getting fired by surprise.
So what happens now? Here’s my breakdown of the fallout or opportunities for the major players involved.
The idea behind OpenAI in the first place was a research organization aimed at protecting humanity from runaway AI. If safety concerns around artificial intelligence are at the heart of this boardroom drama, then directors essentially showed they don’t care that OpenAI is a financial and technological rocket ship.
But now the company faces serious headwinds. OpenAI will have a lot of angry employees who joined thinking they were getting on a rocket ship worth somewhere around $90 billion. It will face major recruiting challenges and fundraising will also become more difficult.
The next generation of AI models will take an untold amount of cash. OpenAI needs to show Microsoft and others that it’s a stable organization hitting on all cylinders. Otherwise, it’s too risky for Microsoft to hitch its wagon to OpenAI.
OpenAI may also need Microsoft more than Microsoft needs OpenAI. There are lots of AI models to choose from and they are getting better every day. There aren’t many cloud compute companies that have invested billions in expanding the infrastructure to train and run AI models.
Still, OpenAI’s customers shouldn’t be meaningfully hurt in the short term, since many already use other models besides OpenAI.
Unless there is some other, more scandalous reason for his departure, Altman and Brockman will likely start a new company and my guess is they will be able to raise massive capital.
Altman is a big name in Silicon Valley. While he’s somewhat polarizing — there are those who bristle at his hard charging, political style — he’ll be able to recruit talent and attract investors.
The software giant agreed to put billions into the company and it still has exclusive rights to use OpenAI models to power its products. But it would be smart of the firm to think about possible replacements now. It won’t be easy but it also has advantages, as OpenAI doesn’t have much without Microsoft, which owns all the compute resources.
Microsoft could use its leverage over the company to push for management changes at the top of OpenAI to ensure it remains a stable partner. But that could end up alienating factions within the company that would like to see it controlled by independent directors who value AI safety above profits. In that sense, it may not be worth Microsoft’s time.
OpenAI’s competitors will be rejoicing tonight. Every major AI company is mobilized now to try and build the biggest, best models and deploy them as quickly as possible. Until Friday, it looked like OpenAI was winning that race by a mile. Now, the picture is very different.
Anthropic, the company founded by a group of former OpenAI employees who also worried about AI safety, could be a real winner after taking a backseat.
Google, which is aggressively competing against Microsoft, is another major victor here. It’s been working to develop an alternative to GPT-4, called Gemini. Google must feel like it just got a second wind.
Amazon is also in a good spot. It has a model-agnostic approach to AI, allowing lots of companies to offer products on the platform. The one model it couldn’t include was GPT-4 because of the exclusive deal with Microsoft. Now, that doesn’t seem as big of a problem.
The bottom line is that no matter what the reason was for Altman’s sudden departure, the leader in the AI race has just been dealt a serious blow and the crown is up for grabs. The game begins now.