Sam Altman Says He Was Wrong: The AI Jobs Apocalypse Never Happened (Yet)

The CEO of OpenAI just told the world he got it wrong on AI and jobs. Speaking at a Commonwealth Bank of Australia conference in Sydney on Tuesday, Sam Altman said he no longer believes in the kind of "jobs apocalypse" some in the AI industry have predicted. Bold words from the man who helped trigger those fears.

Sam Altman admits his AI jobs predictions were off

Altman had previously claimed that AI would "probably replace most of the jobs people do today" and that entire job categories would be "totally, totally gone." Now he says he expected more entry-level white-collar jobs to be eliminated by this point, and his intuitions were "just off."

He shared a personal anecdote: after using AI to respond to his Slack and email messages, labeling replies as "Sam's AI," he realized the human element still matters. People want to interact with people, not bots pretending to be people.

What the data says about AI's impact on jobs

The Yale Budget Lab's econometric analysis found the estimated impact of AI on exposed occupations is close to zero, with the same holding true for real hourly wages. Historically, major technological disruptions play out over decades, not months.

Still, reality is messier than macro data. Meta announced AI-focused job cuts last week, and Cisco confirmed it would fire roughly 4,000 employees in May. Uber's CTO went viral for admitting the company burned through its entire 2026 Claude Code budget in four months.

Convenient timing for OpenAI's trillion-dollar IPO

Here's the elephant in the room. OpenAI confidentially filed its S-1 IPO prospectus with the SEC on May 22, 2026, targeting a valuation of up to $1 trillion. Telling the world that AI won't destroy jobs four days before your IPO paperwork lands is, at minimum, convenient.

Anthropic CEO Dario Amodei said last year that up to half of all entry-level white-collar jobs could dissolve within five years. And a Nvidia VP suggested AI may actually cost companies more than the humans it replaces. Not everyone in the industry shares Altman's newfound optimism.

The AI jobs question remains wide open

Altman is right that the feared collapse hasn't materialized. It is reasonable to expect that widespread effects would take longer than 33 months to show up. The question is whether the apocalypse is cancelled or merely delayed.

Credit to Altman for the honesty. But with a company merging all its products into one super-app and an IPO weeks away, this mea culpa reads as much like PR as philosophy.