Update: Obviously, within 12 hours of this piece being published, Altman was back in at OpenAI (as semi-predicted a few paragraphs down). But that doesn’t change the substance of the piece, of its fundamental question: are we at risk of investing too much of brand equity in single, fallible human CEOs?
What is a CEO?
Most of the people I encounter who have “CEO” on their business cards, lead teams of, I don’t know, 20 or 30 people. They are chiefs amongst a fairly miniscule executive class. But in the wider world — the sort of world I don’t encounter — there’s a different kind of CEO. The kind of CEOs who lead multinational companies with workforces in the thousands. These men — and they’re usually men — are almost all called John. They have greying or white hair, they wear crisp dark blue or charcoal suits which, in order to appear more relaxed at corporate retreats, they pair with a gingham shirt and bulky Asics sneakers. They are pleasant to look at, but not handsome; they are polite, but not nice. They are the titans of American business at the end of the country’s eponymous century.
That is what a CEO is, and that is what a CEO has been for many decades. But a governing principle of this sort of leadership is to avoid the headlines, to stay away from public recognition. If you live in London or New York, Frankfurt or Paris, you’ll likely have walked past a CEO at some point. Brushed against someone on Cheapside who just happens to be one of the most important people in the country. But you wouldn’t know that — John just looks like any old John.
But a new breed of CEO has developed in the technology era, a type of person who defies many of the conventions of CEOdom. They are rarely, if ever, called John. They are predominantly founders, geeky tech whizzes who have been with their company since day dot. They are not corporate luminaries, company men, but entrepreneurs. It is a trend that goes back to Bill Gates founding Microsoft. On paper, Gates now looks like a plausible CEO (“Bill Gates” is, after all, an almost comically boring name), but the clock does not have to be wound back very far before Gates is recognisably less polished, less presentable than his Fortune 500 contemporaries. Remember that video of him dancing at the launch of Windows 95?
I am, of course, not talking about Bill Gates. I’m not even talking about Mark Zuckerberg, the poster child for this sort of new tech CEO who’s attitude was immortalised in Aaron Sorkin’s script for The Social Network (“I’m CEO, bitch,” read the business cards, delivered the day that his relationship with co-founder Eduardo Saverin irreparably deteriorates). In a way, this is a tale of two CEOs: Sam Altman and Elon Musk.
Before it was a multi-billion dollar business at the cutting edge of the next frontier of technology, OpenAI, the company which, until this week, Altman was CEO of, was a not-for-profit founded, in 2015, by a bunch of people including Ilya Sutskever, a Russian-born computer scientist, and Greg Brockman, the former CTO of payments company Stripe. On that initial board were two big characters: Altman and Musk. Altman, aged just 30 at the time, had already been, briefly, CEO of Reddit, as well as a founder of Loopt (sold for $40m+) and president of Y Combinator. Musk, meanwhile, was already Musk.
The impetus for OpenAI was a sense that the development of AI was bound to become one of the most important features of the 21st century, but that, in 2015, it was being left to for-profit and state actors to set the pace for the race. OpenAI received pledges from entrepreneurs like Reid Hoffman, founder of LinkedIn, and Peter Thiel, Musk’s old mucker at PayPal. The mission was to develop “safe and beneficial” artificial intelligence, a mission that continued up to 2019 when the company became a for-profit endeavour (I’m sure that those at OpenAI would claim that the “safe and beneficial” slogan continued after this change, but, ultimately, being of benefit to shareholders and of benefit to society are positions that cannot be held, unproblematically, in tandem).
This week, as you all know by now, Altman was ousted as CEO of OpenAI, a role he’d held since 2019. It was a move that has plunged the company into disarray: it has been reported that 738 of the company’s 770 staff (that’s c.95%) have signed a letter to the board threatening resignation and calling on the board to resign. This is before the substance of the board critique of Altman has even become public domain, and is an extraordinary show of faith from his workforce.
It’s a show of faith that Musk is, perhaps, lacking. This week he finds himself in hot water, once again, after appearing to affirm a dumb conspiracy theory on X. After advertisers, including Apple, pulled away from the service, and CEO Linda Yaccarino was forced to apologise, Musk rowed back a bit, but is now suing a non-profit, Media Matters. The specifics of this incident do matter, but not to this piece. What’s clear is that Musk is continuing to be outspoken, irreverent and public facing. He is, in short, the antithesis of the besuited, sneaker-wearing John-CEOs of the pre-Internet era.
What exactly the OpenAI board’s complaint was about Altman is still unknown, but he, and Brockman, have been offered jobs by Microsoft in the days following their exit. Microsoft, if you don’t know, is OpenAI’s major investor, providing them with $1bn in 2019 and another $10bn in 2023. There are also reports that Microsoft has circulated a generalised offer to OpenAI, claiming they’ll match compensation for anyone looking to jump ship. Microsoft’s 56-year-old CEO Satya Nadella, who was been with the company since 1992 and is considered one of the shrewder players in Silicon Valley, told CNBC that he was “committed to OpenAI and Sam [Altman], irrespective of what configuration.”
It is an interesting, slippery statement. But what’s particularly striking about this whole event is the extent to which Altman has become a commodity. Remember, he’s still just 38-years-old, a Stanford drop-out who has never been engaged in artificial intelligence from a purely computer science perspective (his only, non-board, job at OpenAI has been as CEO). At a raw, almost computational level, his absence from OpenAI will not slow the development of ChatGPT, nor will his presence at Microsoft accelerate their own movements into the sector.
And yet, in whispered circles in the tech community, Altman’s shit-canning by OpenAI is being seen as a potentially existential threat to his former employee. And don’t forget that OpenAI has been THE success story of the past couple of years in tech, the one company who you’d want to back. Since that first release of ChatGPT, OpenAI has felt like Facebook, in its early years, or Apple under Steve Jobs. And now people who know a lot of stuff are suggesting that it might wobble simply due to the absence of a CEO who’s only been in position for 4 years.
But then again, these new CEOs are not John CEOs. Elon Musk can shake Twitter’s share price with a single misjudged tweet. He can also propel a thousand satellites into orbit with just force of imagination. And Altman, who has not been shy in making his views known (“if i start going off, the openai board should go after me for the full value of my shares,” he tweeted on November 18th, in quite a Musklike way) or becoming a brand in his own right. These “brands” comes with positives and negatives for their employers. Clearly, Altman is a superb CEO, inspiring loyalty in his workforce (I suspect that not laying off 80% of them has probably helped) and giving him the greatest contacts book, and hiring potential, in the industry. He may not be able to single-handedly programme the next GPT, but he can recruit the world’s best computer scientists to do so. The difficulty is when the Brand CEO collides with the Brand itself.
In Altman’s case, there has been an increasing divergence between OpenAI as a cautious insurgent in the world of artificial intelligence, and Altman as CEO of the fastest growing company in tech, and the one that every Valley watcher is picking as the next Goliath in the space. Altman has been at the frontline of tech entrepreneurialism since his teens. He was just 19 when he co-founded Loopt, and only 28 when he took over at Y Combinator. He didn’t climb the corporate ladder — he built an elevator and took it straight to the penthouse. And along the way, he encountered managers in a totally new mould — men like Musk and Thiel and Hoffman — and never had to serve under a John CEO, learning at the knee of those grey men in their grey suits.
Musk’s unilateral approach to decision making — anathema to American industrial practices of the past hundred years — has got him into trouble before, notably when he threatened to take Tesla private, or when he, you know, bought Twitter. But it is also perceived as the route of his success. Take away Musk’s ability to make decisions — tough decisions, big decisions — and you will end up with a version of Elon Musk that has all of the flaws and none of the talents. The same is true of Altman: his bravado, that touch of arrogance, has inspired OpenAI to become market leaders. Take that away, grind him down, surround him by cynical MBAs, and what do you have left? Less alt, more man.
It is a delicate balance and one that the tech industry feels increasingly nervous about. Musk might be able to sell cars and launch rockets, but his association with Twitter has been a net negative for the brand (that much now seems clear). What seemed, to his shareholders, like a pleasing dose of capitalist fanaticism now seems like just plain old fanaticism. All the same, there is a lingering feeling, in my mind at least, that if either Musk or Altman had been CEO of Meta these past few years, the Metaverse project might not have withered on the vine. Zuckerberg who, since his early days as the enfant terrible of Palo Alto, has developed a milquetoast statesmanlike demeanour, simply didn’t seem to have the charisma to carry the flame of the future. When he appeared as a legless avatar in his virtual world, the markets got skittish; when Musk sent a Tesla Roadster into space, the markets got excited. The alignment between brand and CEO giveth and it taketh away.
Warming Altman’s old seat at OpenAI is Emmett Shear, who has been drafted in as Interim CEO (this always brings to mind the Red Adair, and red in the face, football managers who are given three months as “Interim Manager” in a bid to keep a team in the Premier League). Shear is a co-founder of Twitch, and served as its CEO from 2011 to 2023. Regular readers of this column will know that I am a fan of Twitch, but it is hard to see what new expertise Shear, with a background in broadcast entertainment technology, will bring to an AI company. But I think there is a decent chance that, by the time you’re reading this, Altman will be safely ensconced back at OpenAI HQ — or that OpenAI itself will be safely ensconced within the warm and welcoming bosom of Microsoft.
Altman’s defenestration is unlikely to spell the end of the Brand CEO era. Likewise, Musk’s ill-advised tweeting on his own platform (the cautionary phrase about “getting high on your own supply” springs to mind) won’t kill off the trend. But there must be board members and shareholders the world over who are getting nervous, not just about what these men might say in a fit of self-importance, but also about the amount of brand equity being invested in single, fallible human beings. If Altman walking away can materially disadvantage OpenAI to the extent that it has, then it stands to reason that any company operating in that manner ought to do its utmost to ensure that no person has that much personal capital.
Humans are flighty, subject to change, error or illness. Brands, on the other hand, are far more resilient. Coca Cola cannot decide it wants to go travelling for a year. Jaguar Land Rover won’t randomly use a racial slur at a gala. Lego isn’t going to come down with a debilitating case of Seasonal Affective Disorder. And the advantage to having a John CEO, rather than a Brand CEO, is that you can shuffle one out and another in, like interchangeable dummies on the great banal production line of life.