Content
summary Summary
Update
  • Added Murati's statement

Update from March 10, 2024:

Ad

Mira Murati comments on the New York Times report via X: The NYT's sources were former board members who wanted to scapegoat her to save their own face in the media, Murati said.

Her relationship with Sam Altman was "strong and productive," and giving clear feedback was part of that. She had been asked by the board for feedback on Altman and had told the board what she had told Altman earlier. She never supported the board's actions.

"I fought the actions aggressively and we all worked to bring Sam back," Murati wrote in an internal message.

Ad
Ad
Image: Mira Murati via X

The investigation into Sam Altman's dismissal appears to be over, with Altman back on the board. What's still unclear is the role of co-founder and chief scientist Ilya Sutskever and whether he will remain with OpenAI. Sutskever was part of the board that fired Altman. He later said he regretted the decision and voted to bring Altman back.

Original article from March 8, 2024:

OpenAI CTO Mira Murati reportedly involved in Sam Altman's firing

OpenAI is still recovering from the drama surrounding the firing and reinstatement of CEO Sam Altman. New details show that the rifts within the company's leadership may be deeper than previously thought.

According to a report by the New York Times, OpenAI CTO Mira Murati played a key role in the temporary ouster of CEO Sam Altman. The NYT's sources say she wrote a private memo to Altman raising questions about his leadership. She also reportedly shared her concerns with the board of directors.

Murati described alleged manipulative behavior by Altman, who would get his colleagues to support his decisions and undermine those who challenged him or, in his view, took too long to decide.

Recommendation

Co-founder and chief scientist Ilya Sutskever also reportedly raised concerns about Altman's manipulative behavior. Both executives described a toxic work environment that Altman had created.

Board members feared that if top executives like Murati and Sutskever left the company because of Altman's behavior, talent would leave.

Instead, hundreds of employees threatened to leave if Altman was not reinstated. OpenAI was on the verge of collapse.

Altman controls OpenAI's AI startup investments

Other concerns included Altman's control of OpenAI's startup fund, which some members felt circumvented the accountability of OpenAI's nonprofit governance structure.

Ad
Ad
Join our community
Join the DECODER community on Discord, Reddit or Twitter - we can't wait to meet you.

The fund, established in late 2021, invests in AI startups and is named OpenAI, but is legally owned by Sam Altman. According to OpenAI, this is a temporary structure to get the fund up and running as quickly as possible.

Until now, Altman's ouster appeared to be a solo effort by the board, largely due to a dispute over a paper co-authored by board member Helen Toner that praised the AI safety approach of OpenAI competitor Anthropic and criticized OpenAI.

Sutskever and Murati have publicly defended Altman. But if the New York Times' sources are to be believed, the rifts in OpenAI's leadership team appear to run deeper.

OpenAI's board of directors is up for renewal in the near future. In addition, the law firm WilmerHale is working on a report on the events, which may bring more details to light. The OpenAI drama might continue.

Support our independent, free-access reporting. Any contribution helps and secures our future. Support now:
Bank transfer
Summary
  • Mira Murati, OpenAI's CTO, and Ilya Sutskever, its chief scientist, raised concerns about CEO Sam Altman's manipulative behavior that led to a toxic work environment, according to a New York Times report. This contributed to his ouster last November.
  • Board members feared a loss of talent if executives like Murati and Sutskever left because of Altman's behavior. However, hundreds of employees threatened to leave OpenAI if Altman was not reinstated.
  • Other concerns included Altman's control of the OpenAI Start-up Fund, which some members say circumvents the accountability of OpenAI's nonprofit governance structure. The fund is legally owned by Sam Altman, but OpenAI says it is only a temporary structure.
Online journalist Matthias is the co-founder and publisher of THE DECODER. He believes that artificial intelligence will fundamentally change the relationship between humans and computers.
Join our community
Join the DECODER community on Discord, Reddit or Twitter - we can't wait to meet you.