OpenAI: Inside the Clash for the World’s Most Important Company

Well, it seems like the epic Silicon Valley drama is finally coming to an end. We finally have a resolution. You haven't been paying attention over the last five days. Let me walk you through really quickly everything that got us to where we are right now. Friday, November 17 OpenAI announces leadership transition. Sam Altman is out as CEO and is also removed from the board. The reason given, Sam Altman was not consistently candid in his communications with the board, hindering its ability to exercise its responsibilities. The board no longer has confidence in his ability to continue leadership leading OpenAI. Vague, but that's what we were working with. We also learn in this statement that Greg Brockman will be stepping down as chairman of the board and will remain in his role at the company. But Greg had other ideas. He immediately quit and gave us a breakdown of the timeline of events, essentially telling us that neither him or Sam were given any prior notice that any of this was about to happen. We also learned that Mira Morati would now be the interim CEO and she found out basically the night before this all happened late Friday night. We then learn that three senior researchers over at OpenAI would then quit in solidarity with Greg Brockman and Sam Altman. We also learned that night that Microsoft and Satya Nadella were also blindsided by this announcement and had no idea this was coming, which is crazy seeing that Microsoft is a 49% investor in OpenAI. Now, the board over at OpenAI consisted of six members. You had Greg Brockman and Sam Altman, and then you had co founder Ilya Suskever, who was the chief scientist for OpenAI. And then we had three people that weren't part of OpenAI on the board. We have Adam D', Angelo, who also is a founder at Quora and creating a chat bot tool called po, which has some competing elements to chat GPT, which for a moment we thought sort of factored into everything that was going on. The board also included Tasha McCauley, a tech entrepreneur, and Helen Toner, somebody that we will come to find out has a little bit of beef with Sam Altman, but we'll get to that in a minute. Based on this board's makeup, they needed four votes to get Sam out. We know Greg didn't vote Sam out. We know Sam didn't vote for himself, meaning that ilya Suskever, Adam D', Angelo, Tasha McCauley and Helen Toner all voted to get Sam out. And remember, Ilya was one of his co founders at OpenAI. Saturday, November 18th we learned that Sam is potentially going to get his job back at OpenAI. After an outpouring of love and support from the employees over at OpenAI and pressure from all of OpenAI's investors, it seemed that Sam was on the fast track to getting his role back and undoing the board's decision. And it appeared that we would have Sam back by the end of the day on Saturday. Throughout the day on Saturday, all of the news looked very optimistic that Sam would be back in his role before the weekend was over. But nope, Sunday we learn that Sam Altman will not be coming back. And instead they brought in the former Twitch co founder Emmett Shear to be the new CEO of OpenAI. So if you're keeping track, that's three CEOs over a three day period. We had Sam Altman, who was fired on Friday. Miram Moratti replaced him. And then by Sunday, Mirmorati was already replaced by Emmett Shear. And then in a huge bombshell, Satya Nadella of Microsoft came and said, fine, if OpenAI isn't bringing Sam Altman and Greg Brockman back, they could come and work for Microsoft. And on top of that, any OpenAI employees that want to come over and work under Sam in this new AI division that we're creating for him, they're all welcome over at Microsoft as well. Sam Altman confirms he's going to Microsoft. Greg Brockman confirms he's going to Microsoft, along with those three researchers. They're that quit in solidarity earlier with them. And then the next twist comes. Ilya Suskever, one of the four board members who voted to get Sam Altman out, went to Twitter to say, I deeply regret my participation in the board's actions. I never intended to harm OpenAI. I love everything we've built together and I will do everything I can to reunite the company. So one of the four people who helped to kick Sam out regretted his decision and now wants Sam and the team back together. Sam Altman sees this, retweets it with three hearts. And we get the impression that maybe there is hope for Sam and Ilya still working together. Monday morning, a letter among OpenAI employees starts to circulate, basically saying, if the board isn't fired, we're all quitting. And one interesting statement from this document was that the board informed the leadership team that allowing the company to be destroyed would be consistent with the mission. The board was ready to let OpenAI die over this feud or whatever was going on here. And notable signers, Miram Moratti, the first person they put as interim CEO to replace Sam Altman and Ilya Suskever, himself, one of the people that voted to get Sam out, signed this document saying that he's going to quit in solidarity if we don't bring Sam back by the end of the day. Over 700 of the 770 OpenAI employees signed this letter. One by one, OpenAI employees took to X to write the same tweet. OpenAI is nothing without its people once again showing solidarity for Sam Altman and bringing the team back together. Emmett she at the time the current CEO of OpenAI, went to Twitter to basically say, I'm going to do an investigation into everything that's going on. I want to know what happened with the board. I want to know the reason Sam was fired. I'm going to get to the bottom of this. Finishing his post with the board, did not remove Sam over any specific disagreement on safety. The reasoning was completely different from that. I'm not crazy enough to take this job without board support for commercializing our awesome models. So essentially reaffirming that he's going to push Forward with the APIs and chat GPT and the commercialization of the and not back off on this stuff. And that's essentially where I left off the story last time I made a video. Now, I did go into a lot of detail around why I think all of this was happening, where the rifts sort of were, and my best guess as to why they decided to get rid of Sam Altman. And again, I think that everybody in the moment thought they were doing the best thing for the future of OpenAI, but nobody totally agreed on what that best thing was. But as we all know, that wasn't the end of the story. We finally got a sort of conclusion by the end of the day, Tuesday night, November 21st. So from my last video to now, here's everything that we learned. We Learned that the OpenAI board at one point approached Anthropic about potentially merging and essentially replacing Sam Altman with the CEO of Anthropic, Dario Amodai. We also learned that before they approached Emmett Shear to be the CEO, they also approached the former GitHub CEO Nat Freeman as well the CEO of Scale AI, Alexander Wang, Modi Friedman and Wang all declined these job offers and eventually they landed on Emmett Shear. Now, while this was all going down, we learned that it was not a done deal yet for Sam Altman over at Microsoft. And are they actually Microsoft employees right now? Like, who do they work for? Yeah. So they're all in the process of joining. And in this Bloomberg interview, Satya also confirms that he has no idea why, why the board removed Sam. No new information was given to him. And when asked the question if he wants changes to the OpenAI board, he essentially said, yes, we need some governance changes over there. Well, as it turns out, Sam was still talking to OpenAI trying to figure out how to come back. We're into Tuesday now and Sam Altman and members of the OpenAI board have opened negotiations aimed at a possible return of the ousted co founder and chief executive officer, the artificial intelligence company. Discussions are happening between Altman and at least one board member, Adam D'. Angelo. The talks also involve some of Open AI's investors who are also pushing for reinstatement. And one thing that came along that was really interesting. Remember I said that bringing in Emmett was probably not the best decision by the board if they really wanted to keep Sam out. Well, the new CEO of OpenAI, Emmett Shear, will quit if OpenAI board can't provide evidence of why they fired Sam. This came out in a report from Bloomberg with Emmett basically saying, I'm not getting any answers about this and I need to get to the bottom of what's going on. At this point, law enforcement was starting to get invol executives said they were getting questions from regulators and law enforcement entities such as the U.S. attorney's office in Manhattan over the charge of Altman's alleged lack of candor. The truth was going to come out one way or another, they told the board. And this is just mind blowing here. The board agreed to discuss the matter with their council. After a few hours, they returned, still unwilling to provide specifics. They said that Altman wasn't candid and often got his way. The board said that Altman had been so deft that they could couldn't even give a specific example. According to the people familiar with the executives. What they're saying here is that the board's claiming that Sam Altman was so deceptive that they can't even give one instance of when he was deceptive because he was that good. Now, of course, the board wasn't saying anything, meaning that the media and people on X and people like me are essentially guessing on the narrative. Without the board speaking, we're all just sort of left speculating. And the early sort of speculation was that Adam d' Angelo was the board member that was essentially keeping Sam out. Adam D' Angelo is the co founder and CEO of Quora.com, but Quora.com also owns the chatbot called Po. And if you're not familiar with Po, Po essentially allows you to create your own custom bots and sell them on a internal bot marketplace. Well, with the release of GPTs and the GPT store, this seemed to be direct competition with PO's Bottom Bot Store, leading to a lot of speculation that Adam d' Angelo was upset that Sam announced the GPT store, which was going to directly compete with his bot store, and that was part of the reason why Adam might have wanted Sam out of the company. More information started to surface about another board member, Helen Toner, who is the director of strategy at Georgetown University center for Security and Emerging Technology. Toner had written a paper called Decoding Intentions, Artificial Intentions, Intelligence, and Costly Signals. The paper came out in October of 23, and it seemed to criticize OpenAI's efforts to keep its AI technology safe. While praising the approach taken by Anthropic, Mr. Altman said that he had reprimanded Ms. Toner for the paper and that it was dangerous to the company, particularly at a time, he added, when the Federal Trade Commission was investigating OpenAI over the data used to build his technology. Toner defended it as an academic paper that analyzed the challenges that the public faces when trying to understand the intentions of the countries and companies developing AI. But Mr. Altman disagreed. He said, I did not feel we're on the same page on the damage of all this. Any amount of criticism from a board member carries a lot of weight. Now, taking a look at this paper in question here, page 30 is where some of the criticism comes into play. You can see here, she says, while the system card itself has been well received among researchers interested in understanding GPT4's risk profile, it appears to have been less successful as a broader signal of OpenAI's commitment to safety. Intended as a relatively inconspicuous research preview, the original Chat GPT was built using a less advanced large language model called GPT 3.5, which was already in widespread use by other AI customers. GPT 3.5's prior circulation is presumably why OpenAI did not feel the need to perform or publish such detailed safety testing in this instance. OpenAI has also drawn criticism for many other safety and ethics issues related to the launches of chat GPT and GPT4, including regarding copyright issues, labor conditions for data annotators, and the susceptibility of their products to jailbreak that allow users to bypass safety controls. This muddled overall picture provides an example of how the messages sent by deliberate signals can be overshadowed by actions that were not designed to reveal intent. She then goes on to praise Anthropic, which happens to be OpenAI's biggest competitor. A different approach to signaling in the private sector comes from anthropic, one of OpenAI's primary competitors. Anthropic's DES to be perceived as a company that values safety shines through across its communications, beginning from its tagline, an AI safety and research company. March 2023 strategy document published on Anthropic's website revealed that the release of Anthropic's chatbot Claude, a competitor to ChatGPT, had been deliberately delayed in order to avoid advancing the rate of AI capabilities progress. Anthropic had deliberately decided not to productize its technology in order to avoid stoking the flames of AI hypothesis hype, where the GPT4 system card acted as a costly signal of OpenAI's emphasis on building safe systems, Anthropic's decision to keep their product off the market was instead a costly signal of restraint. By delaying the release of Claude until other companies put out similar capable products, Anthropic was showing its willingness to avoid exactly the kind of frantic corner cutting that the release of Chat GPT appeared to spur. So basically she's saying that ChatGPT and OpenAI cut corners to get ChatGPT and GPT4 out quickly and Anthropic was doing the right thing by restraining and holding back releasing this to the public before they felt it was safe and ready. Which if you watched my last video about why all of this happened is very consistent with the narrative that I broke down. It was really the story of the people trying to commercialize the product and get it to market as fast as possible for faster commercialization and profitability versus the researchers and the people that were worried about the security and the safety and all of this technology moving too quickly. Now, whether or not Adam d' Angelo was the driving factor that pushed whether it was initially Ilya Suskever, which was the original theory, or whether it was Helen Toner. We don't totally know those details yet, but it seems like each of them had their own sort of separate motives for why it would be a good idea to get Sam out. Fast forward to 1003 Pacific Time. On Tuesday, November 21, OpenAI tweets this we have reached an agreement in principle for Sam Altman to return to OpenAI as CEO with a new initial board of Brett Taylor as the chair, Larry Summers and Adam d' Angelo remaining on the board. So from Friday to Tuesday we finally get a conclusion to this story. Sam Altman was fired. Mira Moratti was in as CEO. Mira turned out to be Team Sam, she's out as CEO. We get Emmett Shear in as CEO, he turns out to be Team Let's Call It Truth. Helps get to the bottom of things and essentially helps broker the deal to bring Sam Altman back on as CEO of OpenAI. So in a five day period, OpenAI has had three different CEOs. Sam Altman goes to Twitter to sort of confirm everything is getting back to normal. I love OpenAI, and everything I've done over the past few days has been in service of keeping this team and its mission together. When I decided to join Microsoft on Sunday evening, it was clear that was the best path for me and the team. With the new board and with Satya's support, I'm looking forward to returning to OpenAI and building our strong partnership with Microsoft. Soft Satya.