What is the AI jobs apocalypse debate? The AI jobs apocalypse debate is the public disagreement between the two leading AI labs, OpenAI and Anthropic, over whether artificial intelligence will eliminate white-collar jobs at large scale or actually expand the work people do. On May 27, 2026, Axios reported that the CEOs are splitting into warring camps, making it nearly impossible for businesses and investors to know what is coming. For commercial real estate investors, this is not an abstract tech argument. Office demand is a direct function of white-collar headcount, so the question of whether AI thins or grows the knowledge workforce sits at the center of every office underwriting model. For the bigger picture on applying AI in your own operations, see our complete guide on the best AI tools for commercial real estate investors.
Key Takeaways
- OpenAI CEO Sam Altman said in May 2026 he was wrong about AI causing a jobs apocalypse, walking back earlier warnings that entire categories of white-collar jobs would disappear.
- Anthropic continues to warn AI could displace human labor at very large scale, though CEO Dario Amodei has softened his prior forecast of 50 percent of entry-level white-collar jobs lost and 10 to 20 percent unemployment.
- On-the-ground data is muted: Stanford and Yale Budget Lab researchers find no significant unemployment shift in high-AI-exposure jobs since 2022, and software engineering openings on Indeed rose 18 percent year over year.
- Both CEOs are softening rhetoric as their companies prepare public offerings that could value each near $1 trillion, so narrative may be doing some of the work.
- Office CRE investors should underwrite demand with scenario analysis, not a single forecast, because even the AI leaders cannot agree on the labor impact.
The AI Jobs Apocalypse Debate Explained
The split crystallized in a pair of public appearances in late May 2026. Speaking virtually to a Commonwealth Bank of Australia conference in Sydney, OpenAI CEO Sam Altman struck an optimistic tone, saying AI is unlikely to cause a jobs apocalypse and that he had been wrong. In his words, he was delighted to be wrong, having expected more elimination of entry-level white-collar jobs by now than has actually occurred. Altman, who previously warned that entire classes of jobs would vanish, said his intuitions about labor displacement were simply off, pointing to his own failed experiment in delegating email and Slack replies to AI before reverting to doing them himself.
Anthropic has held the darker line. At the Vatican's AI ethics conference, co-founder Chris Olah said there is a real possibility that AI will displace human labor at very large scale, echoing CEO Dario Amodei, who once said up to half of entry-level white-collar jobs could dissolve within five years and unemployment could reach 10 to 20 percent. Notably, Amodei has recently begun to hedge, suggesting automation may expand the work people do rather than simply erase it. As Fortune observed, both leaders are walking back their most dire predictions just as their companies prepare blockbuster public offerings, which raises a fair question about whether forecasts or narratives are changing.
What the On-the-Ground Data Actually Shows
For an office investor, the labor data matters more than the rhetoric, and the data so far is reassuringly mixed rather than catastrophic. Unemployment has ticked up since 2023, but predominantly in sectors with the least AI exposure, according to Stanford researchers, which is the opposite of what an AI-driven white-collar collapse would produce. 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. On the demand side, software engineering job openings on Indeed are up more than 18 percent year over year even as total openings fell 4.3 percent, and LinkedIn's chief economist recently attributed roughly 1.3 million new job postings to AI.
That said, the displacement signal is not zero. Tech layoffs through May 2026 passed 115,000, already approaching the 124,000 logged in all of 2025, with Meta, Amazon, and Snap among the firms citing AI as a driver. We covered one of the starkest data points, a CFO survey suggesting AI-attributed cuts run far higher than headline numbers, in our analysis of what the office demand crisis means for CRE investors. The picture, then, is a genuine reshaping of certain roles layered on top of overall labor resilience, not a uniform collapse.
What This Means for Office CRE Demand
Office demand is downstream of two forces that AI pushes in opposite directions. The first is headcount: fewer knowledge workers per firm means less space leased, all else equal. The second is the AI sector itself as a tenant. AI companies have been among the most aggressive office leasers in major markets, a tailwind we detailed in our piece on how AI companies are driving record office leasing in NYC and SF. The net effect on any given submarket depends on which force dominates locally. A market dependent on back-office financial services faces different exposure than one anchored by AI labs and their suppliers.
There is also a subtler dynamic that argues against rapid headcount cuts. Recent research suggests AI productivity is now outrunning the ability of managers to review and approve the work, making management capacity, not raw output, the new bottleneck. We unpacked this in our coverage of why AI is making managers the bottleneck. If judgment and oversight remain scarce human functions, the floor under white-collar employment, and therefore office demand, may be higher than the doom scenario implies. For hands-on help modeling AI's impact on your portfolio's tenant base, CRE investors can reach out to Avi Hacker, J.D. at The AI Consulting Network.
How to Underwrite Office Demand Amid the Uncertainty
When the people building the technology cannot agree on its labor impact, the right response is not to pick a side but to underwrite a range. Practical steps for 2026 office investing:
- Run scenario-based occupancy assumptions: Model a base case, a high-displacement case, and an AI-tenant-tailwind case rather than committing to one forecast. Stress test debt service coverage ratio and net operating income under each.
- Weight tenant industry exposure: Favor assets with diversified or AI-adjacent tenancy over single-industry back-office concentration. Watch lease rollover and weighted average lease term closely.
- Track local leasing data, not national headlines: The AI labor story is highly uneven by market and submarket. CBRE and JLL leasing reports tell you more about a specific asset than any CEO prediction.
- Underwrite shorter holds with optionality: Given the uncertainty, build in flexibility on hold period and capital improvement timing so you can adapt as the labor picture clarifies.
- Discount the rhetoric: Remember that AI leaders have commercial incentives in both directions, especially ahead of major IPOs. Anchor on independent labor data from sources like the Yale Budget Lab and Stanford rather than on founder forecasts.
The AI Consulting Network specializes in helping CRE investors translate fast-moving AI developments into concrete underwriting and portfolio decisions, including office demand scenarios tailored to your specific markets.
Frequently Asked Questions
Q: Why do OpenAI and Anthropic disagree about AI and jobs?
A: OpenAI CEO Sam Altman now says AI is unlikely to cause a jobs apocalypse and that his earlier warnings were wrong, while Anthropic continues to warn AI could displace human labor at very large scale. Both companies are also softening their most extreme claims ahead of large public offerings, so commercial incentives may be influencing the messaging.
Q: Is AI actually eliminating white-collar jobs in 2026?
A: The evidence is mixed. Stanford and Yale Budget Lab researchers find no significant unemployment shift in high-AI-exposure jobs since 2022, and software engineering openings rose 18 percent year over year. At the same time, tech layoffs passed 115,000 in 2026 with some firms citing AI, so certain roles are being reshaped even as overall white-collar employment has held up.
Q: How does the AI jobs debate affect office real estate demand?
A: Office demand tracks white-collar headcount, so large-scale AI displacement would reduce space needs. But AI companies are also major office tenants, creating an offsetting tailwind in markets like New York and San Francisco. The net effect depends on each submarket's tenant mix, which is why local leasing data matters more than national forecasts.
Q: How should office investors underwrite given the uncertainty?
A: Use scenario-based assumptions across base, high-displacement, and AI-tailwind cases rather than a single forecast. Favor diversified or AI-adjacent tenancy, track local CBRE and JLL leasing data, build optionality into hold periods, and anchor on independent labor research rather than AI founder predictions.