OpenAI’s CTO Mira Murati posted on X on Wednesday saying she is leaving the company. Murati said she is stepping away to do her own exploration after more than six years at the AI startup.
“After much reflection, I have made the difficult decision to leave OpenAI,” she said in the post. “There’s never an ideal time to step away from a place one cherishes, yet this moment feels right … My six-and-a-half years with the OpenAI team have been an extraordinary privilege.”
An OpenAI spokesperson declined to comment further, directing TechCrunch to Murati’s tweet.
CEO Sam Altman responded to Murati’s tweet by thanking her in another post. “We’ll say more about the transition plans soon, but for now, I want to take a moment to just feel thanks,” said Altman. “I feel tremendous gratitude towards her for what she has helped us build and accomplish, but I most of all feel personal gratitude towards her for the support and love during all the hard times.”
The decision comes just a week before OpenAI’s DevDay, its annual developer conference.
When Altman was abruptly fired late last year by OpenAI’s previous board of directors, the board briefly installed Murati as interim CEO. Murati was reportedly among those, along with ex-OpenAI chief scientist Ilya Sutskever, who approached the rest of the board prior to Altman’s ouster to express concerns about his behavior.
Altman is increasingly asserting control over OpenAI and its image.
On Monday, Altman penned a blog post saying, among other hyperboles, that OpenAI could achieve “superintelligence” in the next few years. He’s also reportedly poised to receive equity in OpenAI for the first time as the company prepares to move away from its nonprofit governance structure.
Murati came to OpenAI in 2018 as VP of applied AI and partnerships. After being promoted to CTO in 2022, she led the company’s work on the viral AI-powered chatbot ChatGPT, the text-to-image AI DALL-E, and the code-generating system Codex, which powers GitHub’s Copilot product.
She has a degree in mechanical engineering from Dartmouth College and previously worked as an intern at Goldman Sachs and then at Zodiac Aerospace, the French aerospace group. She spent three years at Tesla as a senior product manager of the Model X, the automaker’s crossover SUV, during which Tesla released early versions of Autopilot, its AI-enabled driver-assistance software.
In 2016, Murati joined Leap Motion, a startup building hand- and finger-tracking motion sensors for PCs, as VP of product and engineering. Murati wanted to make the experience of interacting with a computer “as intuitive as playing with a ball,” she told Fast Company in an interview. But she soon realized that the tech, which relied on a VR headset, was too early.
As OpenAI’s CTO, Murati developed a bit of a reputation for making controversial pronouncements.
She once vaguely claimed in an interview that OpenAI’s AI would achieve “Ph.D.-level” intelligence. And in June, Murati raised eyebrows when she suggested that AI would replace creative jobs that “shouldn’t have been there in the first place.”
“Some creative jobs maybe will go away, but maybe they shouldn’t have been there in the first place if the content that comes out of it is not very high quality,” Murati said in an onstage interview at The Wall Street Journal’s WSJ Tech Live Conference. “I really believe that using it as a tool for education [and] creativity will expand our intelligence and creativity and imagination.”
Murati is the latest high-level exec to depart OpenAI in recent months. Sutskever and former safety leader Jan Leike announced their departures in May, and co-founder John Schulman said last month that he was leaving to join rival Anthropic. Meanwhile, OpenAI’s president, Greg Brockman, is on extended leave through the end of the year.
Murati’s decision to step down comes as OpenAI is said to be pursuing a funding round that would value the company at over $150 billion. Microsoft, Nvidia, Apple, and Thrive Capital are reportedly in talks to invest; the round could end up being as large as $6.5 billion, per Bloomberg and others.
OpenAI desperately needs the money. According to The Information, the company has spent approximately $7 billion on model training and $1.5 billion on staffing. At one point in time, ChatGPT alone was said to be costing OpenAI around $700,000 a day to run; Altman has said that training the company’s once-flagship GPT-4 model cost over $100 million.