AI CEOs Walk Back Jobs Apocalypse as Altman Admits 'I Was Wrong'

Two of the most influential CEOs in artificial intelligence spent the last year warning that AI would gut white‑collar employment. Now both are walking those predictions back — just as their companies prepare for blockbuster public offerings that could value each at roughly $1 trillion.

OpenAI CEO Sam Altman, speaking virtually at a Commonwealth Bank of Australia conference in Sydney on Tuesday, said he was "delighted to be wrong" about AI's impact on entry‑level jobs. Anthropic CEO Dario Amodei, who once predicted AI could eliminate 50% of white‑collar roles within five years, has similarly reframed automation as a productivity multiplier rather than a destroyer of work.

The reversals, reported by Fortune, come as both OpenAI and Anthropic eye IPOs this year — a context that has raised eyebrows among skeptics who question whether the timing is coincidental.

"I'm delighted to be wrong about this," Altman told CBA CEO Matt Comyn. "I thought there would have been more impact on entry‑level white‑collar jobs being eliminated by now than has actually happened."

Altman, who previously warned that "entire classes" of jobs would disappear and that AI would "probably replace most of the jobs people do today," now says his intuitions about labor displacement "were just off."

He pointed to a personal experiment as the turning point: Altman tried delegating his Slack and email responses to AI, then found himself returning to manual responses. "We really do care about our interactions with people," he said. "This thing is not something that I can imagine myself outsourcing to an AI anytime soon."

According to TIME, Altman added: "I now think I understand more about why it hasn't, and I'm obviously grateful, but that is an area where my intuitions were just off."

Anthropic CEO Dario Amodei's reversal has been equally striking. In May 2025, he told Axios that AI could wipe out 50% of entry‑level white‑collar jobs and spike unemployment to 10‑20% — a warning he described as a duty: "We, as the producers of this technology, have a duty and an obligation to be honest about what is coming."

Now, as Fortune reported earlier this month, Amodei has reframed automation through the lens of Jevons paradox — the economic principle where increased efficiency leads to increased consumption, not less. "If you automate 90% of the job, then everyone does the 10% of the job," Amodei said. "And the 10% kind of expands to be 100% of what people do and kind of 10‑times their productivity," Amodei said, according to Fortune.

The pivot means automation doesn't reduce demand for workers — it makes their output cheaper and more abundant, potentially expanding the market for their skills.

The timing of both reversals has drawn scrutiny. OpenAI and Anthropic are both reportedly preparing to go public this year, with each company targeting valuations in the $1 trillion range. OpenAI aims to reach $280 billion in revenue by 2030, up from $25 billion today, according to Fortune. Anthropic is reportedly in talks to secure $30 billion in funding at a $900 billion valuation, per the Financial Times.

Peter Wildeford, Head of Policy at the AI Policy Network, told:² "Public opinion research has made pretty clear that Americans feel quite negative about AI. The AI industry, and in particular Sam Altman, has responded with an about‑face: it's not going to replace most white collar workers, it's actually going to create tons of jobs."

He added: "It's hard to say whether they've actually changed their forecasts for AI's economic impact, or whether they're just trying to change the narrative," Wildeford told.²

The evidence paints a mixed picture. Tech layoffs through May 2026 have passed 115,000, according to Layoffs.fyi, already approaching the 124,000 logged in all of 2025. Meta eliminated roughly 8,000 roles this month alone — about 10% of its workforce — citing AI as justification. Intuit cut 3,000 people. AI has been cited for nearly 50,000 job cuts through April, according to outplacement firm Challenger, Gray & Christmas.

Yet the Yale Budget Lab has found no significant changes in occupational mix or unemployment duration in high‑AI‑exposure jobs since ChatGPT launched in late 2022. And Goldman Sachs CEO David Solomon, who never bought the apocalyptic narrative, pointed out in a recent New York Times op‑ed that civilian U.S. employment has grown 145% since 1962, with data center construction alone adding 200,000 jobs since 2022.

"The United States has a long track record of creating new jobs in response to disruption," Solomon wrote. "I don't see any reason to think this dynamic will stop now."

Not everyone in the industry bought the apocalypse narrative to begin with. Nvidia CEO Jensen Huang has been dismissive of what he called the "God complex" among AI executives pushing doom‑laden predictions. In April, he told Fortune that AI will not reduce the number of jobs but will instead create efficiency opportunities for those who adopt the technology.

Nvidia VP of applied deep learning Bryan Catanzaro was more pointed: "For my team, the cost of compute is far beyond the costs of the employees," he told Axios in April.

Box CEO Aaron Levie echoed Solomon's optimism in a LinkedIn post, arguing that automation will increase demand for roles rather than eliminate them: "If you looked at what work looked like a few decades ago and saw how much faster everything is or easier it is to produce today — even before AI — you'd certainly have been convinced there'd be no jobs left. Yet the opposite has happened."

While CEOs walk back their predictions, the ground reality is more complicated. WiseTech, an Australian logistics software company, is cutting one‑third of its workforce — roughly 2,000 jobs — with its new CEO Zubin Appoo declaring that "the era of manually writing code as a core job activity is over." The company had to report "threats of violence" against Appoo to police and increase security at its Sydney headquarters, as reported by the Sydney Morning Herald.

Altman himself acknowledged the tension. He admitted at the CBA conference that despite AI advances, the technology has yet to deliver equivalent economic returns: "The question is, where is the revenue? Where are the actual productivity gains?" Altman admitted at the conference, per the Sydney Morning Herald. "My best answer to that is, it's all still very new, and it's just going to take a little bit longer to figure out how a company actually does run more efficiently," Altman said, according to the Sydney Morning Herald.

He added a note of caution: "If a year from now we're still talking about that same question, I'd be more concerned."

For the developers and builders who actually work with these tools, the CEO reversals validate what many have been saying for months: AI is changing how work gets done, not eliminating the need for human judgment. The tools that builders use — Claude Code, Cursor, Copilot — are accelerators, not replacements.

Altman's Slack experiment is telling: even the CEO of the world's most prominent AI company found that AI couldn't replicate the nuance of human communication. The jobs that are disappearing are the ones that could be scripted before AI existed — and the ones being created are the ones that require judgment, creativity, and the ability to wield AI as a tool rather than be replaced by it.

For builders, the message is clear: the compute might be getting cheaper, but the person who knows how to direct it is becoming more valuable, not less.