Author: Ge Jiaming, Wall Street News

OpenAI suddenly changed its management team in the middle of the night. Sam Altman, who has always been seen as the spokesperson for OpenAI, and Greg Brockman, the former president and core AI scientist known as the father of ChatGPT, unexpectedly left in this "power game"?

OpenAI's statement was "cautious in wording," stating that this decision was made by the board of directors after careful consideration and review, and concluded that Altman was not candid in his communication with the board, hindering his ability to perform his duties. The board no longer believes in his ability to continue leading OpenAI.

"Not candid enough," "hindering the performance of duties," "no longer believed," these three key words can be considered the highest level of "accusations" for a company's co-founder, CEO, and public spokesperson.

This wording quickly sparked various speculations in the industry. Some believe this is a dispute over the distribution of interests as OpenAI enters commercialization, while others believe it involves power struggles among shareholders such as Microsoft.

However, from various signs, the most likely explanation for this "internal struggle" lies in the differences between Sam Altman and Greg Brockman and OpenAI co-founder and chief scientist Ilya Sutskever in AI security, OpenAI's technological development speed, and the commercialization of the company. The reason why Ilya was able to lead the coup with the other three directors is due to the "peculiar" board structure of OpenAI.

Sam Altman and Greg Brockman Speak Out

Just now, Greg Brockman detailed the circumstances of his and Sam Altman's dismissal by the board on social media. He expressed shock and sadness at the actions of the board today. Last night, Sam received a text from Ilya requesting a meeting at noon on Friday.

Ilya told Sam that he would be fired, and the news quickly spread. At 12:23 PM local time, Ilya sent a Google Meet link, informing Greg that he would be removed from the board, and Sam had been fired.

Sam Altman also spoke out for the first time after being dismissed: If I "fire" at the company, the OpenAI board will claim against me.

Just as Sam Altman was fired and Greg Brockman suddenly resigned, three senior researchers at OpenAI resigned on Friday evening local time.

It is reported that the three who resigned are the company's research director Jakub Pachocki, the head of the team responsible for assessing potential risks of artificial intelligence Aleksander Madry, and Szymon Sidor, who has been a researcher at the startup for 7 years.

Internal Struggle in OpenAI's Management

Currently, it is believed that the reason for Altman's departure from OpenAI is likely due to the internal struggle within the management team. The root cause lies in the lack of experience among the senior leadership of the OpenAI board, which has led to recent internal disagreements related to the application of technological breakthroughs, ultimately evolving into an internal struggle.

Analysis suggests that in recent years, due to conflicts of interest, the OpenAI board has lost several senior supervisors, including Tesla CEO Musk, current LinkedIn co-founder Reid Hoffman, and former Republican Congressman Will Hurd. The only person with extensive leadership experience remaining is Adam D'Angelo.

The OpenAI board was previously composed of six people, three OpenAI executives, and three non-employee directors, including: Chairman and President Greg Brockman, Chief Scientist Ilya Sutskever, and CEO Altman, as well as Adam D'Angelo, Tasha McCauley, and Helen Toner.

After Will Hurd's departure, an impasse arose within the OpenAI board between internal executives and the three non-employee directors, and Ilya broke the impasse: previously, Ilya, Sam, Greg, Adam, Tasha, and Helen balanced each other out. However, yesterday, Ilya, along with the other three directors, led a coup, forcing Sam to step down and removing Greg from his position as a director (but retaining Greg's executive role), and Greg subsequently chose to resign on his own.

Media reports revealed that the final vote to oust Altman was 4:2, leading outsiders to believe that the root of this conflict is likely the ideological differences between Ilya Sutskever and Altman.

Therefore, some netizens speculate that firing Sam Altman was not due to his performance, strategic leadership, or vision for the company's future. It was because of controversy surrounding the application of a technological breakthrough, which sparked a voting decision between cautious safety and rapid deployment. Ilya disagreed with Greg and Sam's approach, leading to factional struggles:

If this was pre-planned, Sam would not have represented OpenAI at the APEC event last week, nor would he have attended OpenAI DevDay two weeks ago. If Microsoft had known or been involved, Satya Nadella would not have appeared on stage with Sam at Dev Day.

Microsoft stated that they were not informed in advance - they claim they were only informed a minute before the decision was made. This indicates that there was no consensus internally on this decision, and Greg just resigned for this reason.

At the recent Asia-Pacific Economic Cooperation (APEC) meeting, Sam announced that he had witnessed the boundaries of knowledge being pushed four times during his tenure at OpenAI, with the most recent occurrence happening a few weeks ago.

It seems that Greg and Sam did intend to build and deploy this achievement, but Ilya opposed it. Earlier this year, Ilya took over the leadership of the "superintelligence alignment" department. I believe it was Ilya's vote that broke the original balance, leading to the final disagreement and resignations.

Is the "peculiar" board management structure the root of Altman's dismissal?

So, why was Sam Altman dismissed by Ilya Sutskever? Perhaps the root of all this lies in the "peculiar" board structure of OpenAI.

In 2019, Sam Altman created a groundbreaking organizational structure, consisting of the non-profit OpenAI, Inc. and the for-profit subsidiary OpenAI Global LLC. On the official website, the basic structure of OpenAI shows that a non-profit entity manages a for-profit entity, which can accept investments and set a return rate, but must operate under the governance of the former.

Since the board is still that of a non-profit organization, each director must fulfill their fiduciary duties to advance its mission—safe and broadly beneficial AGI. Although the for-profit subsidiary is allowed to earn and distribute profits, it must adhere to this mission. The primary beneficiaries of this non-profit organization are humans, not OpenAI investors.

This seemingly fair distribution actually harbors a hidden danger, as management from the non-profit entity is usually at the societal level, representing their own values.

The board maintains a majority independence. Independent directors do not hold OpenAI stock, and even Sam Altman does not directly hold stock. His only benefit is indirectly generated through the Y Combinator investment fund, which made a small investment in OpenAI before he worked full-time.

Looking into the backgrounds of these three directors is also interesting:

Adam D'Angelo, most famously known as the co-founder and CEO of Quora, is jokingly referred to by netizens as the person who "messed up" Quora. He primarily focuses on a new AI project—Poe, and it is believed that his interests conflict with those of OpenAI.

Helen Toner is the director of strategy and foundational research funding at the Center for Security and Emerging Technology (CSET) at Georgetown University, providing advice on AI policy and strategy to policymakers and funders.

Tasha McCauley is known for her work in the field of robotics and is the co-founder of Fellow Robots. Her other identity is the wife of the famous actor Joseph Gordon Levitt.

Did Ilya, along with the other three directors, dismiss Altman?

Ilya Sutskever is a co-founder and chief scientist of OpenAI. In July, he formed a new team at the company aimed at controlling the future artificial intelligence systems of "superintelligence."

Before joining OpenAI, Ilya worked at the renowned Google Brain and was a researcher at Stanford University. Media reports citing insiders' information indicated that a month ago, Ilya Sutskever had his responsibilities reduced at the company, reflecting friction between him and Sam Altman and Greg Brockman.

Ilya Sutskever later appealed to the board and won the support of some board members, including Helen Toner, the director of strategy at the Center for Security and Emerging Technology, indicating that he was the mastermind behind this coup.

Media analysis pointed out that the conflict between Sam and Ilya had been brewing for some time. The two individuals, due to their backgrounds, actually had little in common, but their strong personalities, positions, and reputations meant that they were unwilling to yield to each other.

Insider analysis cited by the media indicated that the conflicts between Sam and Ilya included AI security, technological development speed, and the commercialization of the company. Altman's well-known ambitions may have also contributed to this situation:

Altman has been seeking funding from Middle Eastern sovereign wealth funds to create an AI chip startup, competing with processors produced by Nvidia. Altman has also been pursuing investments worth billions of dollars from SoftBank Group Chairman Masayoshi Son. At the same time, Altman has been collaborating with the startup Humane, founded by former Apple designer Imran Chaudhri and Bethany Bongiorno, to develop AI-focused hardware.

The following is a timeline compiled by netizens

November 2: Altman was in the room when the team presented the next major breakthrough. Before this, there had been three previous occasions in OpenAI's history, with the most recent being GPT-4, where they unveiled the unknown and pushed the frontier of discovery. When he saw the latest progress, he was already planning the next steps: fundraising and necessary resources.

November 4: Chief Scientist Ilya Sutskever felt uneasy. Ilya believed they needed more time to figure out the research approach, rather than rushing to deploy the product.

November 6: OpenAI DevDay progressed smoothly, and they would launch the previously released product in the coming days.

After DevDay, Greg and Altman were in full fundraising mode. Their goal was a valuation of $90 billion, three times the $30 billion. OpenAI recruiters called Google employees, promising $10-20 million in compensation if they joined now.

November 9: Greg took a photo with Emmanuel Macron, then attended an investor conference in Paris.

November 13: Altman confirmed for the first time in an interview that work on GPT-5 had begun, and he was fundraising for it.

November 14: Altman paused GPT subscriptions because traffic surged after DevDay, causing the servers to crash.

November 16: At the Asia-Pacific Economic Cooperation CEO Summit, Altman confirmed the next steps on November 4, which made Sutskever angry and scared. The fundraising trip surprised him. The surge in users meant they didn't have enough servers for the search team, which meant they had to raise funds again to support research.

November 17: The board meeting began. Sutskever had employees present progress and risks to the board, which turned into an internal struggle. When he said they shouldn't raise funds and expand, Altman left the meeting without the board's consent.

Altman pointed out that as CEO, he had to be ahead of the team, managing the news flow before the team's needs. He said, "We need more funds to build more data centers."

Sutskever was furious. When things came to a head, the voting results came out. Altman was dismissed, and Greg was kicked off the board.

免责声明：本文章仅代表作者个人观点，不代表本平台的立场和观点。本文章仅供信息分享，不构成对任何人的任何投资建议。用户与作者之间的任何争议，与本平台无关。如网页中刊载的文章或图片涉及侵权，请提供相关的权利证明和身份证明发送邮件到support@aicoin.com，本平台相关工作人员将会进行核查。