How OpenAI So Royally Spoiled Sam Altman’s Shot

New York
CNN
—

OpenAI’s supervisors worried that the company was making the technological equivalent of a nuclear bomb, and its caretaker, Sam Altman, was moving so fast that is at risk of a global catastrophe.

So the board fired him. This might have been the logical solution after all.

But the manner in which Altman was fired—abruptly, opaquely, and without notice to some of OpenAI’s key stakeholders and partners—defied logic. And it risked inflicting more damage than if the board did not take this action.

A company’s board of directors has an obligation, first and foremost, to its shareholders. OpenAI’s largest shareholder is Microsoft, the company that gave Altman & Co. $13 billion to help Bing, Office, Windows and Azure overtake Google and stay ahead of Amazon, IBM and other AI contenders.

However, Microsoft was not informed of Altman’s firing until “right before” the public announcement, according to the CNN contributor. Kara Swisher, who spoke to sources familiar with the board’s ousting of its CEO. Microsoft shares plunged after Altman was let go.

The employees were also not given the news in advance. Neither did Greg Brockman, the company’s co-founder and former chairman in a post on X who learned of Altman’s firing moments before it happened. Brockman, a key supporter of Altman and his strategic leadership of the company, resigned on Friday. Other Altman loyalists also headed for the exits.

Suddenly, OpenAI was in crisis. Reports that Altman and former OpenAI loyalists were about to start their own company risked undoing everything the company had worked so hard to achieve over the past few years.

So a day later, the board reportedly asked for a mulligan and tried to attract Altman. It was a shocking turn of events and an embarrassing standoff by a company that is widely regarded as the most promising producer of the most exciting new technology.

The strange structure of OpenAI’s board complicated things.

The company is a non-profit organization. But Altman, Brockman and chief scientist Ilya Sutskever formed OpenAI LP in 2019, a for-profit entity that exists within the structure of the larger company. This for-profit company took OpenAI from worthless to a $90 billion valuation in just a few years, and Altman is largely seen as the mastermind behind this plan and the key to the company’s success.

However, a company with big backers like Microsoft and venture capital firm Thrive Capital has an obligation to grow its business and make money. Investors want to make sure they are making money and are not known to be patient.

This likely led Altman to push the for-profit company to innovate faster and get products to market. In the grand tradition of Silicon Valley’s “move fast and break things,” these products don’t always work so well at first.

That’s fine, perhaps, when it’s a dating app or social media platform. It’s a completely different thing when it comes to technology that’s so good at imitating human speech and behavior that it can trick people into believing its fake conversations and images are real.

And that’s what reportedly spooked the company’s board of directors, which remained under majority control of the company’s nonprofit wing. Swisher reported that the recent OpenAI developer conference served as a turning point: Altman announced that OpenAI would make tools available for anyone to build their own version of ChatGPT.

For Sutskever and the board, this was a step too far.

According to Altman himself, the company was playing with fire.

When Altman created OpenAI LP four years ago, the new company pointed to his letter who remained “concerned” about AI’s potential to “cause rapid change” for humanity. This could happen unintentionally, with technology performing malicious tasks due to bad code, or intentionally by people subverting AI systems for malicious purposes. So the company was committed to prioritizing safety, even if it meant reducing profits for its stakeholders.

Altman also urged regulators to set limits on AI to prevent people like him from inflicting serious harm on society.

Proponents of AI believe the technology has the potential to revolutionize all industries and improve humanity in the process. It has the potential to improve education, finance, agriculture and health care.

But it also has the potential to put people out of work… 14 million sites could disappear in the next five years, the World Economic Forum warned in April. AI is particularly adept at spreading harmful misinformation. And some, including former OpenAI board member Elon Musk, fear the technology will surpass humanity in intelligence and could wipe out life on the planet.

Not how to handle a crisis

With these threats, real or perceived, it’s no wonder the board was concerned that Altman was moving at too fast a pace. Perhaps they would have felt compelled to get rid of him and replace him with someone who they felt would be more careful with the potentially dangerous technology.

But OpenAI doesn’t work in a vacuum. It has interest groups, some of them with billions poured into the company. And the so-called adults in the room were acting, as Swisher put it: like a “clown car that crashed into a gold mine”, quoting a famous Meta CEO, Mark Zuckerberg, on Twitter.

Involving Microsoft in the decision, informing employees, working with Altman on a dignified exit plan…all of these would have been solutions more commonly employed by a board the size of an OpenAI company, and all with potentially better results.

Microsoft, despite its massive stake, has no seat on OpenAI’s board, due to the company’s odd structure. Now that could change, according to several news reports, including the Wall Street Journal i News from New York. One of the company’s demands, including Altman’s return, is to have a seat at the table.

With OpenAI’s ChatGPT-like capabilities embedded in Bing and other core products, Microsoft believed it had invested wisely in the promising new technology of the future. So it must have come as a shock to CEO Satya Nadella and his crew when they learned of Altman’s firing along with the rest of the world on Friday evening.

The board angered a powerful ally and could be changed forever because of how it handled Altman’s takedown. It could end up with Altman back at the helm, a for-profit company on its non-profit board, and a massive culture change at OpenAI.

Alternatively, it could become a competitor to Altman, who may eventually decide to start a new company and remove talent from OpenAI.

Either way, OpenAI is likely left in a worse position now than it was Friday before Altman was fired. And it was a problem he could have avoided, ironically, by slowing down.