Sam Altman & AI Job Apocalypse, is it real or not?

On May 26, 2026, OpenAI CEO Sam Altman said something that surprised a lot of people who had been taking his earlier warnings seriously.

"I'm delighted to be wrong about this. I thought there would have been more impact on entry-level white-collar jobs being eliminated by now than has actually happened."

This was a direct reversal. In prior years, Altman warned that AI would eliminate large categories of entry-level white-collar work, that "jobs are definitely going to go away," and that entire job categories would be "totally, totally gone." He is now saying he got that wrong.

Moreover, Dario Amodei, Anthropic's CEO emphasized that AI may expand and reshape work in some areas, although he continues to warn that significant labor-market disruption remains possible.

Two of the most prominent AI doomsayers are somewhat reversing course in the same week. Worth reading carefully.

One detail worth reading twice

Both OpenAI and Anthropic are eyeing large IPOs. OpenAI filed IPO paperwork confidentially in May 2026. A calmer narrative around job displacement is significantly better for a public listing than an active panic about automation wiping out the workforce.

That doesn't automatically make the reversal wrong. But earlier in 2026, OpenAI published a policy paper calling for taxes on automated labor and pilots of a 32-hour working week. That paper explicitly assumed significant labor market disruption is coming. It is difficult to read that document and Altman's "delighted to be wrong" interview as coming from the same position.

Altman acknowledged the gap himself: "At the highest level, we've been roughly right on technological predictions and pretty wrong on the social and economic implications."

What is the Data saying?

Tech layoffs in 2026 passed 142,000 by late May, already exceeding the 124k logged in all of 2025. Meta, Amazon, Microsoft, and Snap are among the companies cutting at scale. Several cited AI as a driver, with internal memos describing workforce reductions as enabling "the substantial investments we are making" in AI infrastructure. Big Tech collectively is spending $725 billion on AI capex in 2026. Some of that math does require a smaller headcount.

At the same time, some analyses (including work from the Yale Budget Lab) report no meaningful change in unemployment rates through March 2026, including for workers in high AI-exposure roles. These findings depend on the dataset and timeframe examined, so layoffs are real even if broad unemployment disruption has not yet materialized.

A BCG report from April 2026 offers the clearest framing. After analyzing roughly 165 million U.S. jobs, BCG found that 50 to 55% of U.S. jobs will be substantially changed by AI over the next two to three years. Substantially changed is not the same as eliminated. BCG projects 10 to 15% of roles could be fully eliminated over five years. The rest will be reshaped: same title, different task mix, different expectations for what fills the hours.

Pro tip: The question "will AI take my job?" is less useful than: "Which tasks in my role are already automatable, and am I the one running that automation or the one being replaced by it?" Those are very different positions to be in, and they are not determined by your job title.

What practitioners are saying

The ground-level experience matches the BCG framing more than the apocalypse framing.

What practitioners are saying: In 2026 Quora and Reddit threads on AI and white-collar work, the recurring theme is task-level change rather than wholesale displacement. A frequently referenced example: a customer support agent no longer handles every ticket manually. They supervise multiple AI assistants and escalate what the assistants cannot resolve. The role still exists. What fills the hours has completely changed. Workers who adapted early report higher output and in some cases stronger job security. Workers who resisted or were not supported through the shift are the ones being squeezed. The pattern is consistent: adapt the task mix, keep the role. Don't adapt, get exposed.

This matches what Anthropic's own research found in March 2026: AI models were being used for only a "fraction" of the work-related tasks they are already capable of handling. The technology is ahead of adoption. Most of the labor market has not been reorganized around it yet.

What this means for your team

Whether the apocalypse arrives turns out to be the wrong question for most operators.

The more useful question: is your team building skill in directing AI, or are they spending hours on tasks that AI already handles competently?

Companies cutting at scale in 2026 are not replacing humans with AI uniformly. They are shifting toward smaller teams that multiply output using AI tools and reducing headcount that does not. That is a different threat model than mass displacement, and it calls for a different response.

Small teams that adopt AI-powered workflows are not replacing their people. They are choosing which tasks get human time and which get automated. Output stays the same or improves; manual drag drops. That is the actual productivity story.

Author note: The 142K layoffs are real people, and I don't want to paper over that. But the right response for most operators is not to wait for the prediction market to settle. It is to audit the recurring workflows on your team and figure out which ones shouldn't require human hours. That gap is where the productivity difference compounds over the next 12 months.

That is exactly where a brand agent changes the equation. A SketricGen brand agent works on your behalf around the clock — handling customer questions, qualifying leads, and routing support — without adding to your headcount. You define how it behaves, what it knows, and where it shows up: your website, WhatsApp, Slack, or anywhere your customers already are. The tasks get handled; the payroll stays flat. That is what it looks like to multiply a small team without hiring your way through a disruption.

The window to adapt is open. The debate over whether the apocalypse is coming is a distraction from a simpler question: are you getting more from your current team than you were six months ago?

Try SketricGen and deploy your first brand agent without writing a line of code. Or start from a template to see what others have already built.

Sources & further reading

FAQs

Both, depending on the role. BCG's April 2026 analysis of 165 million U.S. jobs found that 50 to 55% will be substantially reshaped over the next two to three years. A separate 10 to 15% of roles are projected to be eliminated within five years. The consistent pattern: highly repetitive, rules-based tasks are being automated first. Roles that require judgment, client interaction, or oversight of AI systems are the ones holding.

Altman said on May 26, 2026 that he underestimated the "human part" of work and the importance people place on interacting with others on the job. He acknowledged being "pretty wrong on the social and economic implications" of AI, even while maintaining his technical predictions were roughly accurate. The reversal came days after OpenAI filed IPO paperwork confidentially, which is relevant context given that a calmer jobs narrative is considerably better for a public listing.

Readiness matters more than worry. The teams feeling squeezed in 2026 are largely the ones that did not adapt, where manual tasks kept accumulating while competitors automated them. BCG's analysis specifically identifies a "Divergent Roles" category where junior positions shrink while oversight and senior roles grow. Building AI literacy on your team now is a lower-risk posture than waiting to see where the disruption lands.

Start at the task level. Identify the customer-facing and internal workflows that are repetitive, rule-based, and time-consuming: answering the same questions, qualifying inbound leads, routing support requests, scheduling. A brand agent handles all of that without adding headcount. SketricGen lets you build and deploy one from a plain-English description — no code required. Starting from a template is the fastest way to get a working agent in front of customers before building something custom.

Related blogs

View more
How to Build and Use MCPs (Model Context Protocol) in 2026

How to Build and Use MCPs (Model Context Protocol) in 2026

ai trends and research
The Rise of AI Workflow Platforms: Why Businesses Are Moving Beyond Traditional Automation

The Rise of AI Workflow Platforms: Why Businesses Are Moving Beyond Traditional Automation

ai trends and research
Multi-Agent AI Explained: How AI Systems Work Together

Multi-Agent AI Explained: How AI Systems Work Together

ai trends and research
What Is an AI Workflow Builder? (And Why Every Business Needs One in 2026)

What Is an AI Workflow Builder? (And Why Every Business Needs One in 2026)

ai trends and research
AI Agents in 2026: Trends, Stats, and Real-World Impact

AI Agents in 2026: Trends, Stats, and Real-World Impact

ai trends and research