The TRUTH About Sam Altman's Firing; Peter Thiel, Eliezer Yudkowsky and Effective Altruism

Okay, here's a breakdown of the wild ride that was Sam Altman's firing from OpenAI, based on the video.

This video dives into the behind-the-scenes drama surrounding Sam Altman's temporary ousting from OpenAI in late 2023, exploring surprising connections between tech bigwigs like Peter Thiel, AI safety proponents like Eliezer Yudkowsky, and the Effective Altruism movement. It paints a picture of boardroom conflict fueled by differing philosophies on AI safety and rapid development, ultimately leading to a corporate showdown and Altman's eventual return.

Here are the key points and juicy details:

• A new book is coming out that apparently spills more details on the OpenAI drama from November 2023, providing some previously unheard insights.
• Surprisingly, Peter Thiel was involved in the story. He's a major figure in the tech world, a mentor, part of the "PayPal mafia" with Elon Musk, and runs the Thiel Fellowship, which funds promising young people who skip college to build companies.
• The Thiel Fellowship has an amazing track record of finding talent, including Vitalik Buterin (Ethereum founder), Dylan Field (Figma co-founder), and Austin Russell (Luminar founder), many of whom became billionaires.
• Peter Thiel also mentored or mentored Sam Altman. In fact, in November 2023, right around the firing, Thiel and Altman were having dinner.
• During this dinner, Peter Thiel warned Sam Altman about Eliezer Yudkowsky, another AI-focused person Thiel had funded (specifically Yudkowsky's AI safety institute).
• Eliezer Yudkowsky is considered the original AI alignment person and a key figure in the "AI doomer" community. He's famous for his interviews (like with Lex Friedman and Dwarkesh Patel) and founded the Less Wrong website.
• Yudkowsky's core belief, summed up simply, is that superintelligence will "just kill you." He's really focused on the extreme risks of developing advanced AI.
• Yudkowsky also plays a big role in the Effective Altruism (EA) movement. EA recently shifted focus from things like ending global poverty to preventing catastrophic risks from AI.
• Peter Thiel warned Sam Altman that Yudkowsky's ideas had "programmed half the people in your company," meaning many OpenAI employees were influenced by the extreme AI safety/risk concerns associated with EA.
• Thiel predicted that the EAs within OpenAI would "end up destroying OpenAI." Sam Altman reportedly responded by comparing them to Elon Musk, who was also removed from the board, saying they "got rid of Elon." This suggests there were existing tensions or a perceived problem with certain individuals on the board.
• At the time of the firing drama, OpenAI's board consisted of Ilya Sutskever, Helen Toner, Adam D'Angelo, Tasha McCauley, Greg Brockman, and Sam Altman. The board controls the non-profit entity at the top of OpenAI's complex structure.
• Helen Toner and Tasha McCauley are apparently associated with the EA movement.
• While Sam and Greg weren't present, the other four board members (Ilya, Helen, Adam, Tasha) held secret meetings to discuss firing Sam Altman.
• Reasons cited for wanting to fire Sam included alleged "safety breaches."
• One supposed breach was that Sam claimed a joint safety board with Microsoft approved all OpenAI releases, but a board member (Toner) allegedly found proof only one was approved.
• Another alleged safety issue was launching GPT-4 earlier in India for testing without the joint safety board's approval.
• Issues with the OpenAI startup fund were also a factor. The board was reportedly unaware Sam Altman personally owned the fund, even though it was said to be a temporary tax arrangement and he didn't earn fees from it. It was seen as an "unusual arrangement."
• The massive, unexpected success of ChatGPT's launch (intended as a "research preview") was also a point of tension, as some board members felt Sam wasn't transparent or didn't handle it appropriately from a safety perspective, despite Sam/Greg later saying nobody could have predicted its impact.
• It seems Ilya Sutskever might have initiated the chain of events, subtly suggesting to Helen Toner that she should talk to Mira Murati (OpenAI's CTO) for more information.
• There were rumors of conflict between Ilya Sutskever and another researcher, Jakob Pachocki, potentially over who would lead OpenAI's research direction, with Sam Altman allegedly promising the role to both.
• Helen Toner had published a paper criticizing OpenAI's safety efforts compared to Anthropic's, which reportedly upset Sam Altman, who then allegedly told Ilya that Tasha McCauley thought Toner should be fired. The board members reportedly realized this was a lie during their secret meetings.
• The four board members decided to fire Sam when he was out of town, calling him on a video call.
• Greg Brockman was also removed from the board, likely because Mira Murati, who they wanted to step in as interim CEO, didn't want to report to him.
• In a bizarre twist, after the firing, the remaining board members allegedly told Ilya Sutskever they suspected he might have been sent as a "spy" by Sam Altman to test their loyalty and instigate the events, which shows the level of paranoia and distrust.
• This whole saga, the presenter believes, is absolutely material for a movie or TV series, highlighting the boardroom drama, alleged double agents, and the "cult-like" moment of Ilya chanting "feel the AGI" with employees.
• Public opinion polls from the time showed most people felt the handling of the situation was "inept," with a large majority thinking Helen Toner either had bad intentions or executed poorly.
• The board handled the aftermath poorly, failing to inform key stakeholders like Microsoft CEO Satya Nadella beforehand (he found out just before the news broke on a Friday) and providing only a vague explanation ("not consistently candid").
• They expected Mira Murati, the interim CEO, to manage employees and keep things calm while they figured out the company's future (potentially even selling it to Anthropic).
• Instead, Mira Murati led a company-wide revolt against the board. A narrative spread within the company blaming Ilya's potential conflict with Pachocki and Helen Toner's perceived retaliation for Sam disliking her paper.
• Because the board didn't provide a clear explanation, employees and the public filled the vacuum with their own theories and frustrations.
• By Monday morning, almost all OpenAI employees, including Mira Murati and surprisingly Ilya Sutskever, signed a letter threatening to quit and join Microsoft (who publicly offered them jobs) unless Sam Altman was reinstated.
• This pressure worked. Sam Altman was brought back as CEO. Helen Toner and Tasha McCauley were removed from the board. Adam D'Angelo remained on the board.
• The EA-associated board members were gone shortly after.
• Since then, Ilya Sutskever left OpenAI to start his own safe AI company (Safe Superintelligence), and Mira Murati reportedly also left and started her own company (though the video incorrectly states this, as she is still CTO at OpenAI as of late 2023/early 2024).
• The presenter is curious about the interpretation: Was it a failed coup by EA-aligned members genuinely concerned about safety but fumbling the execution? Was it something more "nefarious" to install someone aligned with their views? Or was it potentially a "double cross" orchestrated, perhaps with Peter Thiel's influence, to get the EA-aligned board members removed?

Overall, the video emphasizes the intense ideological clashes and personal conflicts at the top of a leading AI company, suggesting the Sam Altman firing was a culmination of tensions involving different views on AI risk, governance, and potentially internal power struggles, all with surprising connections to prominent figures like Peter Thiel and movements like Effective Altruism.