Sam Altman’s removal as OpenAI CEO set off a chain of resignations and left employees, investors and company leaders seeking answers. The board said Altman had not been consistently candid in communications with it, but the reasons behind that claim were not made clear. Reports pointed to a broader disagreement over the pace of the company’s growth and technology deployment.
A sudden decision with immediate consequences
Altman was asked on Thursday night to join a remote board meeting at noon the next day. Greg Brockman, then chairman of the board, was not invited. After Altman was fired, Chief Scientist Ilya Sutskever told Brockman that he was being removed from his board role but could stay at the company. Brockman declined and resigned later on Friday.
OpenAI’s management team learned of the changes shortly after they happened, according to Brockman’s account. Former CTO Mira Murati had been informed on Thursday night and became interim CEO. The abrupt announcement also surprised Microsoft, a key investor and minority owner. Microsoft CEO Satya Nadella was reportedly furious.
The departures extended beyond the two executives. Three researchers resigned on Friday: Jakub Pachocki, GPT-4 lead and OpenAI’s director of research; Aleksander Madry, head of a team evaluating AI risk; and Szymon Sidor, an open source baselines researcher. Their exits added to uncertainty about how the dispute would affect research and product work.
A disagreement over speed and direction
People described as insiders said tensions had grown over Altman’s push for commercialization and company growth. Sutskever, they said, argued for slowing down. One reported turning point was OpenAI’s Dev Day event on November 6, where Altman presented consumer-like products in a keynote. Sources described the event as a moment when he had pushed too far, too fast.
That account places the board’s action in the context of a disagreement about how to balance progress, business growth and safety. OpenAI’s structure makes that balance especially consequential: its for-profit arm is owned and controlled by a nonprofit 501(c)(3) public charity. The nonprofit board’s stated duty was tied to the mission of ensuring that OpenAI builds AGI that benefits all of humanity.
Before Altman’s removal, the board included Altman, Brockman, Sutskever and three members who were not OpenAI employees: Adam D’Angelo, Tasha McCauley and Helen Toner. Afterward, Sutskever, D’Angelo, McCauley and Toner remained. The change left a smaller board facing questions about how it reached its decision and how it would explain it.
Safety concerns and unanswered questions
OpenAI’s pursuit of artificial general intelligence, or AGI, is central to the dispute described in the reports. AGI is a hypothetical technology able to perform any intellectual task a human can do. Sutskever co-led OpenAI’s Superalignment team, which was tasked with figuring out how to control hypothetical superintelligent AI.
Altman had recently spoken about a new advance at OpenAI, saying it was one of several times he had seen the frontier of discovery move forward. The source article cautioned that the concern was not necessarily that OpenAI had developed superintelligence. Instead, a breakthrough could have added pressure to an existing divide: the nonprofit’s safety mission alongside the commercial goals of the for-profit subsidiary.
The board’s public explanation raised a separate question. Its announcement said Altman had not been consistently candid in his communications with the board, hindering its ability to exercise its responsibilities. The company had not clarified that accusation. OpenAI COO Brad Lightcap’s internal memo, obtained by Axios, said the decision was not made in response to malfeasance or issues involving financial, business, safety, or security and privacy practices. He described it as a breakdown in communication between Altman and the board.
What the company said about moving forward
Altman and Brockman said they were shocked and saddened by the board’s action. Others inside the company emphasized that work would continue. OpenAI engineering manager Evan Morikawa wrote that the company would keep shipping and that its work came from people across research, product, engineering and design.
Lightcap told employees the announcement had surprised the company and said he was trying to understand the board’s reasoning and process. He also voiced concern about how the process had been handled, expressed support for Murati as interim CEO and said the company was working toward resolution and clarity.
The reports offered competing signals: a board decision linked to its responsibilities, a communication breakdown cited by the COO, and accounts of a power struggle over leadership and pace. With the board’s accusation still unexplained, the central questions remained how the decision was made and whether OpenAI’s leaders could resolve their differences while pursuing both safety and growth.