Five AI Leaders Spoke About Jobs This Week. The...

Key Takeaways

Five major executives spoke publicly about AI and job displacement between May 21 and 26, each reaching a different conclusion.
David Solomon (Goldman Sachs) called job fears "overblown" and said 25% automation equals reallocation, not elimination. Jamie Dimon (JPMorgan) disagreed, expecting fewer bankers and more AI specialists through attrition. Both are probably right at their own scale.
Sam Altman (OpenAI) admitted on May 26 that he was "pretty wrong" about the pace of white-collar displacement. In January, Dario Amodei (Anthropic) warned of 10–20% unemployment risk. By May he'd reversed: automate 90% of a job and demand for the remaining 10% grows fast enough to expand total employment.
Jensen Huang (Nvidia) said the lazy take is that AI replaces people. His version: people who use AI will replace people who don't. Uber's COO noted that AI cost-justification is still murky in production, even with heavy adoption.
The disagreement is real, not performative. The people most informed about AI scaling can't agree on labor market impact. That's the signal. For practitioners, the answer isn't in waiting for consensus.

The same week, five voices, five positions

Between May 21 and 26, the CEOs and top executives most directly invested in AI all went public with employment forecasts. None agreed. That part isn't surprising. What is: the people who actually build these systems, who track cost curves, deployment rates and adoption friction, couldn't reach consensus on the most basic question. Does AI eliminate jobs or create them?

David Solomon, Goldman Sachs CEO, went first. In a May 22 New York Times op-ed, he argued that fears about AI-driven job losses are "overblown." Automation handles 25% of finance tasks, which means reallocation and upskilling, not elimination. New roles emerge. The historical pattern holds. Solomon's argument: your firm is running the same playbook banks ran with electronic trading.

Jamie Dimon, JPMorgan CEO, disagreed in the same news cycle. Speaking on May 21 and 22, Dimon said AI "will reduce our jobs down the road." Fewer bankers required. More AI specialists hired to replace them. Job loss happens through attrition and the hiring mix shifts. That's not elimination. It's transformation.

Jensen Huang rejected the entire framing. The lazy narrative is that AI replaces people. Huang's version: "You won't lose your job to AI. You'll lose it to someone who uses AI." That's a competitive pressure argument, not a technology argument. The question isn't whether AI is coming. It's whether you get there before your competitors do.

Then the people who actually build the AI changed their minds.

The reversals

In January 2026, Dario Amodei, CEO of Anthropic, issued a stark warning. AI would displace 10–20% of white-collar workers. The pain would be "unusually painful." The window between 2026 and 2031 would be the hardest transition. Amodei wasn't being rhetorical. He was modeling the impact and warning the world.

By May, four months later, Amodei reversed. He flipped to an older economics idea: make something 90% more efficient and demand for the remaining 10% grows fast enough that total employment actually expands. Economists call it the Jevons Paradox. People still do the work. The lever just got longer. Amodei's January apocalypse became a demand-expansion story by spring. Both analyses came from the same data. The January forecast was wrong.

Sam Altman went further. On May 26, he posted that he was "delighted to be wrong" about white-collar job displacement. He'd predicted faster, deeper impact. The reality has been slower. He even tested GPT on his own Slack and dropped it. The tool didn't improve his workflow enough to justify the mental overhead. The man who built the most widely-used AI just admitted it's less disruptive than he thought.

What changed between January and May? Not the technology. The technology got better. What changed is how messy the real-world deployment turned out to be. AI integrates harder than anyone predicted. The ROI isn't automatic. Displacement is running slower than the January math suggested. The two people who sounded the loudest alarm have both walked it back. That should get your attention.

The operating reality check

Uber's COO added something the CEOs weren't saying. AI costs are "hard to justify" inside operating companies. Token usage doesn't translate cleanly to useful output. The automation is real. The value extraction is inconsistent. Your vendor can show you a 40% task automation rate and it still won't tell you whether to deploy it.

What actually matters for your firm

Nobody here is wrong. Solomon and Dimon are describing the same situation from different vantage points. Scale and labor mix determine whether you see reallocation or reduction. Huang is describing competitive reality: the firms that adopt early eat the margin of the ones that don't. Altman and Amodei just admitted the pace is slower than they forecast. Uber's COO is dealing with the same ROI problem you are.

The signal for a CPA firm isn't to wait for consensus from Fortune 500 CEOs. The move is your own firm-level decision: what work to automate, how to hire and what to charge clients when your cost structure shifts.

Gartner's models show net job creation from AI beginning in 2028. The World Economic Forum projects 78 million net new jobs by 2030 after 92 million are displaced. Both are probably true at the global macro level. At your firm level, the relevant questions are:

Which of my current workflows do I want to automate in the next 18 months?
What does my labor model look like if I do that?
Will I hire specialists to replace the displaced volume or upskill existing staff?
How does this affect my pricing and client engagement model?
What am I willing to sacrifice in service breadth to move faster on execution?

None of those are technology questions. They're strategy questions. Goldman, JPMorgan, OpenAI and Anthropic can't answer them for you. Your own leadership needs to.

The CEOs are still debating whether it's 2028 or 2032 when everything changes. Your firm's decision day is now.

The real signal

Altman and Amodei both reversed course. That's worth sitting with. We're 18 months into GPT-4 scale and the two most informed voices have stopped predicting with confidence. Not because they're suddenly optimistic. Because the picture is genuinely murky. Stop waiting for their consensus. Make the call for your own firm.

Frequently Asked Questions

Did Goldman Sachs and JPMorgan actually disagree about AI job losses?

Yes, and both were describing the same market from different vantage points. David Solomon said fears are overblown and 25% automation means reallocation. Jamie Dimon said fewer bankers will be needed over time. Both can be right depending on firm size and labor mix.

Why did Sam Altman say he was wrong about AI job displacement?

Altman predicted faster and deeper white-collar displacement than has actually happened. He tested GPT on his own Slack and dropped it because it didn't improve his workflow enough. He called the pace "slower than expected" in May 2026.

What is the Jevons Paradox and how does it apply to AI jobs?

The Jevons Paradox says that making something more efficient usually increases total demand rather than reducing it. Dario Amodei used it to argue that automating 90% of a job can grow demand for the remaining 10% fast enough that total employment expands.

What does Gartner say about AI job creation?

Gartner's models show net job creation from AI beginning in 2028. The World Economic Forum projects 78 million net new jobs by 2030 after 92 million are displaced. Both figures describe the global macro level, not what happens at any individual firm.

What questions should a CPA firm ask about AI and staffing?

The practical questions are firm-level: which workflows to automate in the next 18 months, whether to hire specialists or upskill existing staff, how automation affects pricing and what service breadth to sacrifice to move faster. These are strategy questions, not technology questions.

Five AI Leaders Spoke About Jobs This Week. They All Said Something Different.