- A failed coup at OpenAI saw Sam Altman return as CEO, and stronger than ever.
- OpenAI staffers displayed enthusiastic loyalty to Altman after his firing last week.
- Those who know him point to his charisma and communication skills.
A board firing a powerful, apparently successful CEO is rare in Silicon Valley.
That CEO coming back after just five days is unheard of.
The botched coup of Sam Altman, the CEO of OpenAI, has seemingly reinforced his power. None of the reasons floated for his firing — alleged dishonesty, alleged disregard for AI safety, or good old-fashioned internal politics — were serious enough to keep him out.
Even more astonishingly, Altman lacks the one obvious lever of power that keeps founder-CEOs such as Mark Zuckerberg in power: controlling shares in his company — Altman owns no shares in OpenAI.
How has he done it?
“He’s super charismatic and super smart,” Eduard Cristea, the CEO of the real-estate-sales startup Holofy, told Business Insider. “Almost immediately, you kind of feel it.”
That charisma was evident from the wave of ❤️ emoji that flooded X over the weekend as OpenAI staffers expressed solidarity and outrage at Altman’s firing.
Hundreds then posted the phrase: “OpenAI is nothing without its people” on X and signed a letter to the board threatening to quit unless it reinstated Altman.
Another indicator is that when Satya Nadella, the CEO of Microsoft, moved at lightning speed to hire Altman and Greg Brockman, Altman’s cofounder who followed him to the exit, the tech giant’s stock price hit a record high.
‘Master of persuasion’
Vinod Khosla, a venture capitalist and an investor in OpenAI, labeled Altman a “once-in-a-generation CEO.” Dan Siroker, the CEO of Rewind AI, said he was “one of the true missionaries” in a world full of mercenaries in Business Insider’s April profile of Altman.
Paul Graham, the founder of the famed Silicon Valley startup accelerator Y Combinator who appointed Altman as his successor in 2014, described him as a “master of AI and persuasion,” ahead of last week’s fallout.
Cristea said he became known as an “incredible guide” and as someone who was keen to “pay it forward” to those who worked with him at Y Combinator.
“The thing that is really impressive is his communication style,” Cristea said. “He’s really, really, really good at both being understood, but also understanding and asking thoughtful questions.”
Those who have interacted with Altman praise his “straightforward communication” and “incredible curiosity” dating back to his Stanford days.
It’s clear, too, that Altman was able to rally OpenAI’s staff around the higher purpose of creating artificial general intelligence, or AGI, the unrealized idea that AI can outsmart humanity.
“It’s one of these really intense environments where people are working extremely hard,” one former OpenAI staffer said of the company.
“They’re there on weekends, they’re there evenings, and there are a lot of people who are either obsessed with OpenAI because they’re obsessed with startups, or they’re obsessed with the actual mission of creating AGI.”
The staffer said some people who worked alongside Altman before OpenAI’s transformative launch of ChatGPT in 2022 would “die in the trenches with him.”
“It’s not the case that around the lunch table, everyone will want to talk about Sam and how brilliant he is or something like this,” they said.
“I get the feeling that what happened is that these senior loyalists were making calls through the night to get everyone else to sign the letter.”
It’s all about the money, money, money
In the run-up to Altman’s ousting, OpenAI was reportedly exploring a share sale that would have allowed employees to cash in on their stakes. And The Information reported that it would’ve been a lot of cash: The deal was reportedly set to value OpenAI at $86 billion.
It’s possible employees mobilized to support Altman to avoid jeopardizing that deal and their payouts.
“Everyone knew that their money was tied to Sam. They acted in their best interest,” one OpenAI staffer posted on Blind, a social-network site where employees can post anonymously.
“Even recent employees would have seen a 4x cash out at liquidation. If the company drastically slowed commercialization, then no investor would’ve come in.”
OpenAI’s future looks tied to Altman
OpenAI has an unusual corporate structure.
When Altman and his cofounders set OpenAI up in 2015, they did so as a nonprofit with the aim of building safe artificial intelligence “for the benefit of humanity.” Three years later it added a “capped profit” company called OpenAI LP. This corporate arm brought in cold hard cash and computing power — but therein lay a tension.
OpenAI’s nonprofit board was installed to ensure OpenAI stayed true to its original mission and ensure the benefit of mankind came ahead of commercial ambition.
But Altman is, fundamentally, an entrepreneur. And last week’s chaos implies his commercially driven vision could win out.
The board that ousted him has been largely deconstructed. Ilya Sutskever, the chief scientist and a cofounder of OpenAI who backed Altman’s firing, publicly expressed his regret at his involvement.
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.
— Ilya Sutskever (@ilyasut) November 20, 2023
One theory is that Sutskever, who represents OpenAI’s research side, and Altman were divided on whether the startup was prioritizing making money over responsibly developing AI, The New York Times reported.
“I think what’s happened over the last few years is that Ilya has been really worrying now that the technology is coming of age and actually works,” the former staffer said. “That would have been what would have driven this wedge between them.”
Another former staffer said raising concerns at OpenAI did not always go well. In Business Insider’s April profile on Altman, one ex-employee said the prevailing mentality was: “If we listened to you, we’d never get the product out.”
For Altman, the board will largely be new, but the workforce will be immensely familiar.
“I love OpenAI,” he posted after his return on Wednesday.
OpenAI seems to love him back.
Read the full article here