Among AI’s big dogs, doomerism is so out, and boomerism has boomeranged back.
Last year, Sam Altman warned that AI would make whole categories of jobs disappear. Anthropic CEO Dario Amodei said half of white collar entry-level jobs could vanish in five years. Palantir CEO Alex Karp has gone so far to say that the only people sure to have a place in the AI era will work in trades or be neurodivergent. Even famed workaholic Elon Musk envisions a world where AI strikes desk jobs “like lightning.”
That apocalyptic messaging captivated public attention. It became part of the rationale for near-trillion dollar valuations, it sold enterprise subscription packages, and it gave employers cover to slash jobs by the thousands and pivot to AI. But the rhetoric became very unpopular. It made many fearful that more wealth would concentrate among those in charge of AI systems and shove the rest of us into a “permanent underclass” as our skills become obsolete.
The job guillotine view no longer serves AI companies. Now rosier messaging has arrived — and just in time as they chase historic IPOs (Anthropic filed a confidential S-1 yesterday).
Last week, Altman said he was “delighted to be wrong” about past predictions that AI would swallow white-collar work, particularly entry-level jobs. Earlier in May, Amodei recalibrated his AI warnings, painting a picture instead where AI can supercharge productivity. His cofounder Chris Olah, meanwhile, visited the Vatican and said, “There is a real possibility that AI will displace human labor at very large scale. If that happens, supporting those displaced will be a moral imperative of historic proportion.” Just weeks after saying “trauma and disruption” were coming, Musk said future work would be optional, likening it to a hobbyist gardner who chooses to grow their own vegetables when they could buy them at the store — a future that certainly sounds more appealing than being struck by lightning.
The whiplash may do more to confuse than calm people who have been riled up by their past warnings.
“They’re trying to get ahead of it, but personally I think they’re already behind,” says Eric Fischgrund, founder and CEO of FischTank PR, which works with tech companies. When it comes to saving AI’s reputation, “it’s too late for the general public right now, unless some of these organizations start making a very real and transparent effort to provide some sort of education or literacy around AI or start training everyday people.”
Every time Amanda publishes a story, you’ll get an alert straight to your inbox!
Stay connected to Amanda and get more of their work as it publishes.
Bob Hutchins, an AI strategy advisor to companies, nonprofits, and educational organizations, agrees. “You can’t go to the public market selling societal collapse,” he says. The AI CEOs “spent 2025 speaking to a tech press audience that rewards big claims, but in 2026 they’re speaking to bankers, retail investors, and a public that’s tired. The audience has changed. They’re pushing back, and so the message is following the environment.”
You can’t go to the public market selling societal collapse.Bob Hutchins, an AI strategy advisor
The hard rhetorical pivots comes as public opinion on the tech plummets. Respondents to a March poll from NBC had such negative view of AI, it had the third lowest net positive ratings of all topics (-20). Only the Democratic party (-22) and Iran (-53) were less popular. Grassroots groups have mobilized to block data center development, graduates boo commencement speakers who champion AI, and a man is facing charges for allegedly throwing a Molotov cocktail at Altman’s home in April. A recent Gallup poll found Gen Z feels increasingly anxious and angry about AI. Employees everywhere are agitated at AI usage mandates as they wade through the meaningless slop generated by their coworkers, work more intensely the more AI tools they use, and burn out.
Even if the job obliteration warnings were in good faith, they ran up against a workforce plagued with anxiety and existential dread around employment. ChatGPT debuted and the job takeover claims came as the tech industry laid off thousands of people, rightsizing itself after years of overhiring. So when warnings of further job losses came from on high, they hit a nerve with fed up workers. The messaging also gave companies cover to say they had slashed thousands of jobs due to AI, as if it was the inevitable path forward to surviving a massive workforce overhaul. The truth is stickier. Many companies have shifted money spent on salaries to spend on AI investments, and about half of those who previously cut jobs in customer service and cited AI as the rationale have plans to rehire for such roles. Altman has also pushed back against this characterization, saying many layoffs blamed on AI would have happened with or without the tech. While unemployment rates have risen for recent graduates, the overall unemployment rate has ticked up only slightly since April of 2024, from about 3.9% to 4.3%, and there’s no evidence of massive job displacement taking hold.
The narrative around AI has seesawed wildly for the almost four years. The launch of ChatGPT launched thousands of thinkpieces about the end of writing and work as we knew it; while others claimed that increased efficiency would lead to more, not less, demand and consumption (i.e., the reason your social feeds were flooded with references to Jevons paradox). Then AI execs began sounding the apocalyptic alarm bells, and others followed their lead. Ford CEO Jim Farley said last year that half white collar jobs would vanish; Fiverr CEO Micha Kaufman wrote to staff in a memo that “AI is coming for you.”As they change their tone again, other companies may in turn shift their messaging, but workers won’t forget what they’ve heard about the risks to their livelihoods.
It’s important to not take everything AI execs say as the definitive AI future. Amodei created a stir earlier this year when he said he couldn’t rule out if AI models could be conscious (they are not). Beneath the cataclysmic warnings and punchy soundbites, both Altman and Amodei have hedged their comments about the job apocalypse, noting one thing that is certain: We do not know the extent to which AI will impact the workforce and society. AI has not replaced vast swaths of workers or driven massive productivity gains across industries. For now, its true power lies somewhere at the intersection of over-hype, apocalyptic threats, and use cases that demonstrably democratize skills and empower people to do more.
But uncertainty isn’t easy to sell — not to regulators, the public market, or people hugging their jobs for dear life. To clean up its reputation, AI needs to consistently prove useful to the average person.
Amanda Hoover is a senior correspondent at Business Insider covering the tech industry. She writes about the biggest tech companies and trends.
Business Insider’s Discourse stories provide perspectives on the day’s most pressing issues, informed by analysis, reporting, and expertise.
