Sam Altman's tech villain arc is underway

Sam Altman appears to be entering a new era. Recent exits from OpenAI's safety team and a dispute with Scarlett Johansson have brought scrutiny.

• OpenAI CEO Sam Altman appears to be entering a new era.
• Altman was lauded as the leader behind ChatGPT when it launched in 2022.
• But recent exits from OpenAI's safety team and a dispute with Scarlett Johansson have brought scrutiny.

Many of them start the same, with a potent mix of genius and idealism and a promise to improve the world with their brilliance. People believe them — pouring millions of dollars into their nebulous ideas — and soon, they are gracing the covers of magazines and headlining summits and conferences the world over.

And then, inevitably, gravity does its thing. They fall.

We've seen it time and again with tech founders: Mark Zuckerberg went from boy genius to a string of scandals. Elon Musk went from "the real Tony Stark" out to save the world to critics casting him as the caricature of an evil billionaire seeking world domination. Elizabeth Holmes and Sam Bankman-Fried were once regarded as saviors. Now, both are in prison.

It looks like Sam Altman, who has been hit with bad headline after bad headline over the past couple of weeks, is the latest to succumb to the narrative: The OpenAI CEO and cofounder appears to have entered his villain era.

After ChatGPT took the world by storm in late 2022, Altman went from being virtually unknown outside Silicon Valley circles to a household name. He was named Time Magazine's 2023 CEO of the year, and Microsoft showed its confidence in him with a $10 billion investment. His short-lived ousting last year — marked by colleagues and the tech elite rallying around him — only strengthened his power (though it also pointed at some of the AI safety concerns that have flared up again this week.)

"It's Sam's world, one prominent tech developer told BI last year. "And we're all living in it."

All the while, he promised to do good through his work.

"I think we can have a much, much better world," Altman told BI last year about AI's potential. "I think there's a few things that need to happen for it. And I like feeling useful."

Meanwhile, Altman was personally investing in clean energy and research into antiaging, and he even cut a check to help a startup fund payroll during the collapse of Silicon Valley Bank.

"Sam Altman is a hero of mine," former Google CEO Eric Schmidt tweeted in November "I, and billions of people, will benefit from his future work — it's going to be simply incredible. Thank you @sama for all you have done for all of us."

From hero to antihero: the initial villain inklings

As Bruce Wayne once learned — and Altman is likely learning in real time — you either die a hero or live long enough to become the villain.

For months now, there have been murmurs that Altman may not belong on the pedestal the world was putting him on. (To be fair, Altman has never claimed to be perfect or all-knowing and has been quick to respond directly to criticism.)

Some in his circles have raised eyebrows at his web of investments, which seemed to promote an outlook reminiscent of a sci-fi film.

"Look, if your worldview is that you have AGI and it's basically superhuman, and these are like the gods, we invented god, yeah then maybe you should turn the whole planet upside down," Databricks CEO Ali Ghodsi told BI earlier this year. "I don't think that's what's happening."

There were signs in March that some in the VC world were starting to see past the Altman hype.

"Him and his brother have always been superhyped," one VC partner told BI in March, referring to Sam and his brother Jack. "It's always been like, 'Oh, the Altman brothers,' it's just going to be way overpriced just because of who they are."

"He's one of the more intellectually dishonest guys in tech," another said at the time."I've had plenty of meetings with him where he says things where I'm like, 'That just cannot possibly be true,' but he can kinda get away with it."

Sam Altman's rough couple of weeks

Now, the criticisms seem to have expanded beyond Silicon Valley circles.

Last week, OpenAI announced its new flagship AI model, GPT-4o, which can interact via text, audio, and video. On X, Altman seemed to liken the new model to the AI from the movie "Her" — but it's not clear he watched the film to its dystopian completion.

Critics were quick to respond, saying the AI assistant sounded sexualized and was too flirtatious. One said it gave them "mega ick," and others said it sounded eerily similar to Scarlett Johannson, who voiced the robot in "Her."

The actor released a scathing statement, saying she had declined multiple offers from Altman to voice the AI and was "shocked" and "angered" by the similarity.

A week after the voice, named Sky, was launched, it was put on pause. The company has maintained it did not deliberately set out to mimic Johannson's voice.

Still, the seeds of a dystopian future — one in which artificial intelligence could take on a human's likeness without their permission — were sowed.

The company was simultaneously waging a defensive war on another front. The day after GPT-4o was released, cofounder and chief scientist Ilya Sutskever and machine learning researcher Jan Leike stepped down. Together, they led OpenAI's superalignment team, which had seen several other departures in recent months.

By the end of the week, the team, whose job it was to keep humans safe from AI superintelligence, was completely disbanded, leaving OpenAI's commitment to safety in question.

"Over the past years, safety culture and processes have taken a backseat to shiny products," Leike wrote in a series of social media posts announcing his departure.

To make matters worse, Vox published a damning report about OpenAI and Altman's leadership. Former employees were unhappy with safety standards at the company and lost their trust in Altman, per the report, and they were afraid to talk about it because they could lose vested equity if they disparaged the company or even mentioned signing NDAs.

Altman appeared caught with his back against the ropes. He and OpenAI president Greg Brockman wrote a lengthy social media post on Saturday delineating their commitment to safety.

In response to the question of the unusual NDAs, Altman expressed regret: "This is on me and one of the few times I've been genuinely embarrassed running openai; I did not know this was happening, and I should have."

But he would only look worse. In a follow-up, Vox reported that it seemed as if Altman did know about the equity.

If tech's fallen star history repeats itself, things are not looking good for Altman. His dream of a utopia seems more like a dystopia to some, and his image as a Silicon Valley do-gooder is unraveling.

Of course, it's not all over for Altman. As far as we know, he's done no crime that would see him suffer the same fate as Bankman-Fried and Holmes.

And there's the chance he can pull a rebrand — something no one has done better than Zuckerberg. It looks like Altman may need to hit the gym, pick a few well-timed fights with another tech billionaire, and maybe even get a chain necklace.