‘Not consistently candid’: The cryptic phrase that ended Sam Altman’s tenure at OpenAI

It was a bombshell that rocked the tech world. Sam Altman, the CEO of OpenAI, the most prestigious and visionary AI startup on the planet, was sacked by the board on Friday. And he wasn’t the only one. Greg Brockman, the organization’s president, was also ousted. It was a swift and unexpected move, one that raised many questions and doubts. 

But the explanation that the board gave for their decisive action was even more baffling. They said that Altman was “not consistently candid in his communications with the board and the broader OpenAI team.”

Not consistently candid. What a strange and cryptic phrase. What did it imply? What did it conceal? Why did the board throw Sam under the bus so publicly?

The phrase “not consistently candid” is the kind of euphemism that makes you wonder what the hell is going on behind the scenes. It’s vague enough to imply something serious, but not specific enough to reveal anything concrete. It’s a way of saying “we don’t trust this guy, but we don’t want to tell you why.” It’s a phrase that begs for more explanation, more context.

That’s why it’s perhaps the most important detail from the shocking ouster of Sam Altman at OpenAI. But what does that mean, exactly? What did Altman lie about, or omit, or distort, or exaggerate?

Was it about his vision and strategy for OpenAI, which was founded as a nonprofit research lab with a lofty mission to create artificial general intelligence that would benefit humanity, but later became a hybrid entity that also pursued commercial interests and partnered with Microsoft?

Was it about his rumored venture fund that he was starting, which could pose a conflict of interest with his role at OpenAI and his access to its research and talent?  Was it about a possible data leak from the launch of the GPT Store at OpenAI Dev Day, which could expose sensitive information and intellectual property of both OpenAI and its partners? Or was it about a new startup he was planning to launch, which could compete with OpenAI or Microsoft in the AI market?

Or was it something else entirely, something more personal, more scandalous? The rumor mill was spinning over the weekend, as some sources claimed that Altman was planning to return to OpenAI after some intervention from Microsoft CEO Satya Nadella and other influential figures. And then there was the bombshell accusation from his sister, Annie Altman, who alleged on social media that he had sexually assaulted her when they were teenagers.

By Sunday night, it was clear that Altman’s fate was sealed. He was not coming back to OpenAI, and instead, he was hired by Microsoft to lead a new advanced AI research team. The board announced that Emmett Shear, the former CEO of Twitch, would take over as the interim CEO of OpenAI. They also expressed their gratitude to Altman and Brockman for their contributions, and their commitment to OpenAI’s mission and values.

But the gratitude and the commitment rung hollow, given the abrupt and dramatic nature of the leadership transition. And the mission and the values seemed murky, given the lack of transparency and accountability from the board.

And the phrase “not consistently candid” remains a mystery, a puzzle, a riddle for the public to stew on. It’s a phrase that demands more scrutiny, more investigation as we learn more about exactly what happened. It’s a phrase that matters, not only for the future of OpenAI, but for the future of AI itself. In other words: What the hell did Sam Altman allegedly lie to the board about?