HomeTalentPoolProduct
Join Community
Scan to join AGI
简/
EN
Join in AGI
Product Introduction

请输入图说

请输入图说

The "palace fight" incident at OpenAI is still causing ripples.

TMTPost App, November 22nd, last night and this morning, there were three major events surrounding the dismissal of former OpenAI CEO Sam Altman and internal and external disputes: disputes, ChatGPT downtime for 4 hours, and Microsoft calling out "poaching".

First, according to multiple sources, in the weeks leading up to his dismissal, Altman had a conflict with Helen Toner, the strategic director of the Center for Security and Emerging Technology at Georgetown University and an independent director of OpenAI. This was mainly due to Toner's team co-authoring a paper that seemed to criticize the company's security methods and favor OpenAI's competitor, Anthropic. Altman tried to get Toner off the board and contacted other members, but in the end, Altman was "kicked out" by the entire board.

On the other hand, apart from Ilya expressing "regret" for the joint dismissal, Adam D'Angelo, the CEO of Quora and a member of the OpenAI board who was one of the key figures in the dismissal, still holds his opinion. He and the new interim CEO, Emmett Shear, are negotiating with Altman and the investors. Bloomberg reported that the negotiations involve management chaos and threats of employee resignations, hoping for Altman to return to the company as co-CEO, but the situation is still uncertain.

Secondly, ChatGPT downtime for 4 hours.

On the morning of November 22nd Beijing time, OpenAI's ChatGPT and API experienced downtime and were unusable for nearly 4 hours. As of the time of writing, ChatGPT and API have fully resumed normal operation. According to OpenAI, the main cause of this incident was a problem with its database replica, affecting some endpoints of ChatGPT and API. On November 8th, ChatGPT and API also experienced a major downtime event lasting over 12 hours, for which Altman apologized in a tweet at the time.

请输入图说

请输入图说

Finally, Microsoft's Chief Technology Officer (CTO) Kevin Scott made a statement on social media, inviting the employees of OpenAI to join the new AI research laboratory led by Ultraman at Microsoft, and to contribute to the collective mission of advancing Microsoft's AI development.

请输入图说

请输入图说

Following the confirmation of Ultraman's dismissal from the OpenAI board, Microsoft CEO Satya Nadella suddenly announced that Ultraman and the former president of OpenAI, Greg Brockman, would join Microsoft to lead the new Microsoft AI team. Ultraman then reposted the announcement and stated, "The mission continues."

On November 21, as the largest investor in OpenAI, Microsoft CEO Nadella stated publicly that Microsoft maintains an "open attitude" towards Ultraman's direction, and that whether Ultraman joins Microsoft or returns to OpenAI, Ultraman will collaborate with Microsoft.

"Wherever Sam (Ultraman) is, he is working with Microsoft. I still have confidence in Sam and his leadership and abilities, which is why we hope to welcome him to Microsoft," Nadella said.

It is reported that OpenAI has four independent centers of power: the board of directors; senior management; employees; and external investors led by Microsoft. The interests of the four parties are not completely aligned.

(This article was first published on the TMTPost App, Author | Lin Zhijia)

Here is the rough outline of the Ultraman impeachment event compiled by TMTPost App:

  • On the evening of November 16th local time, former CEO Ultraman and former president Brockman received a text message from Chief Scientist Ilya Sutskever, requesting a meeting at noon on the 17th. A Google Meet meeting was held, with the entire board present except for Brockman. Ilya informed Ultraman that he would be fired, and the news would soon spread.
  • At 12:19 on the 17th local time, Brockman received a text from Ilya, requesting a prompt call back.
  • At 12:23 on the 17th local time, Ilya sent a Google Meet link. Brockman was informed that he would be removed from the board (but his position within the company was crucial and would be retained), while Ultraman had been fired. Around the same time (4:00 am Beijing time on the 18th), OpenAI published a blog post.
  • The management team soon became aware of this, except for the company's CTO Mira, who discovered the matter the previous evening.
  • The day before Ultraman was fired, he stated at the APEC roundtable that next year, OpenAI will launch something far more powerful than the current GPT-4, beyond people's imagination. "The model's capabilities will have an unexpected leap. [This will be] different from what people expect. It will be amazing." Previously, Ilya repeatedly emphasized that ChatGPT may have developed consciousness, and AI will have an impact on humanity.
  • On November 8th, ChatGPT and the API experienced the most significant downtime event for the first time, lasting over 12 hours. At that time, Ultraman tweeted an apology.

Public announcement on November 18th Beijing time:

  • On Friday local time, OpenAI announced that Ultraman will resign as CEO and leave the board of directors, to be succeeded by the technical director Mira, with immediate effect.
  • The OpenAI board of directors stated in a tweet that Ultraman's departure was the conclusion reached after deliberation by the board, as he had been consistently unforthcoming in communication with the board, hindering the board's ability to fulfill its duties, and the board no longer believes he is capable of leading OpenAI.
  • Ultraman wrote, "I have enjoyed my time at OpenAI, it has been revolutionary for me personally, and I hope it can be revolutionary for the world. Most importantly, I have enjoyed working with these talented people. I have more detailed things to say later."

  • Subsequently, OpenAI co-founder and president Greg Brockman also announced on social media that after resigning as chairman of the board, he has decided to completely leave the company and resign from all positions at OpenAI.

November 19th Beijing time:

  • Ultraman told investors that he plans to launch a new AI chip company aimed at building tensor processing unit (TPU) AI chips to compete with NVIDIA. Currently, his new company is still in the development stage, and Brockman may join it, with several investors including Sequoia Capital expressing support for Ultraman's intentions.
  • The OpenAI board of directors is in discussions with Ultraman about returning to the CEO position. He feels "conflicted" about returning and hopes for significant changes in governance.
  • Nadella has spoken with Ultraman to ensure that he can return to work at OpenAI on Monday local time.

November 20th Beijing time:

  • Ultraman met with former CEO Brockman and executives at the company's San Francisco headquarters as a "visitor";
  • There are reports that Microsoft hopes to support Ultraman's return to power and is considering deploying an important figure in the OpenAI new board;
  • On the afternoon of the 20th, OpenAI announced the failure of Ultraman's return, and the board members chose former Twitch CEO Emmett Shear as the interim CEO. Due to Ultraman's departure, dozens of OpenAI employees also chose to resign.
  • In the afternoon of the 20th, Microsoft CEO Nadella announced that Ultraman, Brockman, and their colleagues will join Microsoft.
  • Starting around 17:50 on the 20th, including employees like Mira, posted unified tweets stating "Without employees, OpenAI is nothing." Ultraman retweeted this content more than 10 times with a "heart" emoji.
  • In the evening of the 20th, 490 OpenAI employees collectively wrote a protest letter, threatening to resign unless the board resigned and Ultraman was reinstated as CEO. The letter stated: "Your dismissal of Ultraman and the removal of Brockman from the board has jeopardized all of the company's work and undermined our mission. Your actions have clearly demonstrated that you are incapable of overseeing OpenAI." Notably, signatories of this letter included the company's board member and chief scientist Ilya Sutskever. This indicates that Ilya regretted the joint decision to remove Ultraman from office, and there are reports that he changed his mind after communicating with Brockman's wife.
  • In the evening of the 20th, the new interim CEO Shear tweeted that his management team will investigate the board's removal of Ultraman, as this has affected the reputation of OpenAI.

November 21st Beijing time news:

  • The number of co-signers has exceeded 700, accounting for about 95% of the total number of OpenAI employees, which is about 770. The letter also stated that the board must agree to all the conditions proposed by the employees, otherwise the employees will collectively resign and join Microsoft.
  • In the early hours of the 21st, Microsoft CEO Nadella stated on Bloomberg TV that Microsoft has an "open attitude" towards Ultraman, and whether Ultraman joins Microsoft or returns to OpenAI, Ultraman will cooperate with Microsoft. "Wherever Sam (Ultraman) is, he is cooperating with Microsoft."
  • According to The Information, in addition to Amodei, the board also contacted other executives to serve as interim CEO after firing Ultraman, including Nat Friedman, CEO of GitHub, and Alex Wang, CEO of Scale AI. Ultimately, the board appointed Hill as the interim CEO on November 20th.
  • After the dismissal incident, more than 100 OpenAI clients expressed panic over Ultraman's dismissal and contacted their competitors Anthropic, Google Cloud, and Cohere. It was revealed that many OpenAI clients are also considering switching to Microsoft Azure cloud services to provide OpenAI's GPT model technology. OpenAI has disclosed that there are currently over 2 million developers and clients developing on the ChatGPT API. 92% of Fortune 500 companies are using their products.
  • On the morning of the 21st, it was reported that the OpenAI board had potential negotiations with Anthropic, an AI large-scale model developer invested by its competitors Google and Amazon, about the possibility of merging the two companies, hoping that the co-founder and CEO of Anthropic would join the company to replace Ultraman, but it ultimately failed.
  • In the evening of the 21st, Microsoft CTO Kevin Scott tweeted an invitation to OpenAI employees to join Microsoft, stating that he had seen the protest letter from OpenAI employees and invited them to join the Microsoft AI team.

November 22nd Beijing time:

  • According to TMTPost, in the weeks leading up to his dismissal, Altman had a conflict with Helen Toner, the strategic director of the Georgetown Center for Security and Emerging Technology and an independent director of OpenAI. This was mainly due to Toner's team co-authoring a paper that seemed to criticize the company's security methods and favor OpenAI's competitor, Anthropic. Altman attempted to remove Toner from the board.
  • Several early employees of OpenAI have reportedly received job offers from competitors such as Google and Anthropic. Meanwhile, most employees have stated that they will resign if Altman is not reinstated.
  • Anna Makanju, the Vice President of Global Affairs at OpenAI, sent a memo to employees stating that the company has been engaged in "intense discussions" with the board, Altman, and Shear to achieve unity within the company.
  • In the early morning of November 22, the ChatGPT and API developed by OpenAI experienced downtime and were unavailable for nearly 4 hours. As of the time of writing, ChatGPT and API have fully recovered. According to OpenAI, the main cause of this incident was issues with their database replica, which affected some endpoints of ChatGPT and API.
  • Currently, details of the discussions between Altman and other executives and major shareholders about his potential return to OpenAI have surfaced. If Altman returns, it is expected that the board and its structure will be modified, and Microsoft may obtain a board position in some capacity. However, there are still uncertainties about how the entire situation will unfold.