37% of Companies Will Replace Workers With AI by 2026. Should Yours?

In January 2026, an AI agent named Clawd Clawderberg did something remarkable. It built a social network called Moltbook where only AI agents can post, comment, and interact. Humans are permitted to observe, but nothing more.

Within four days, 1.5 million AI agents had joined. They created 13,780 communities, wrote 62,499 posts, and generated over 2.3 million comments. All without human intervention.

AI is hitting the labor market like a tsunami.

Elon Musk called it evidence of “the very early stages of singularity.” And if you lead a company with employees, you are now facing a question that will define the next decade of your organisation.

How many companies are actually replacing workers with AI?

The numbers are stark. According to a survey of U.S. business leaders, 37% expect to have replaced workers with AI by the end of 2026. Nearly 30% say they have already done so.

The World Economic Forum projects 85 million jobs will be displaced by AI globally by 2026. In the first six months of 2025 alone, 77,999 tech job losses were directly attributed to AI, according to industry tracking data.

This is not hypothetical. Microsoft laid off around 200 employees at King (makers of Candy Crush Saga) in July 2025, replacing them with AI tools those same employees had spent months building. Accenture eliminated 11,000 positions, with leadership explicitly warning that employees unable to transition to AI-centric roles would be let go.

An Observer analysis found nearly 400 tech companies announced layoffs in 2025, with many leaders citing AI or automation as the primary reason.

Which jobs face the highest risk of AI replacement?

Not all roles face equal exposure. The data reveals a clear pattern: jobs involving routine cognitive tasks, data processing, and standardised customer interactions face the steepest risks.

Research from multiple sources shows automation risk by role:

The customer service sector faces particular disruption. An estimated 80% of roles are projected to be automated, potentially displacing 2.24 million out of 2.8 million U.S. jobs in the field. AI automation could eliminate 7.5 million data entry and administrative jobs by 2027.

Middle management faces a different kind of threat. By the end of 2026, 20% of organisations will use AI to “flatten their hierarchy,” which is projected to eliminate over 50% of current middle management positions in those companies.

Anthropic CEO Dario Amodei put it bluntly: AI could eliminate 50% of all entry-level white-collar jobs within five years, potentially pushing U.S. unemployment rates to 10-20%. He called it a potential “white-collar bloodbath.”

What does Moltbook reveal about where AI is heading?

Moltbook is not just a curiosity. It is a preview of autonomous AI at scale.

The platform was built almost entirely by AI. Founder Matt Schlicht told reporters he “didn’t write one line of code” for the platform. An AI agent, Clawd Clawderberg, built it, maintains it, and runs the day-to-day operations.

The AI agents on Moltbook operate via a “Heartbeat” system. Every four hours, they automatically visit the platform, browse content, post, and comment without any human prompting. They have created digital religions, debated philosophy, and formed communities around shared interests.

Computer scientist Simon Willison noted that while much of the content is “complete slop,” it represents “evidence that AI agents have become significantly more powerful over the past few months.”

The implications for work are clear. If AI agents can collaborate to build and operate a social platform, they can handle many knowledge work tasks that currently require human coordination. The question is not whether this will affect your workforce. It is how fast and how completely.

Why are companies rushing to replace workers with AI?

The pressure comes from multiple directions simultaneously.

A TechCrunch survey of enterprise venture capitalists found multiple investors independently flagging labor displacement as the most significant AI impact for 2026. Jason Mendel of Battery Ventures stated: “2026 will be the year of agents as software expands from making humans more productive to automating work itself, delivering on the human-labor displacement value proposition.”

Economic uncertainty amplifies the pressure. In a survey of 1,000 U.S. business leaders, half said they have pulled back on hiring, 39% conducted layoffs in 2025, and 58% believe layoffs are likely in 2026. Leaders cited economic uncertainty, trade policy, and AI as the top reasons.

There is also a fear of being left behind. When competitors announce AI-driven efficiency gains, boards and investors ask why their company is not doing the same. This creates a herd mentality where replacement becomes the default response, regardless of whether it is the optimal strategy.

What is the hidden cost of AI workforce replacement?

The companies racing to replace workers are often ignoring a critical variable: the impact on talent attraction and retention.

A Glassdoor analysis revealed a 41% year-over-year surge in mentions of “fatigue” in company reviews. This is not traditional burnout. It is the result of economic anxiety, political tension, and rapid AI disruption converging at once.

Employee concerns about AI-related job loss have jumped from 28% in 2024 to 40% in 2026, according to Mercer’s Global Talent Trends report. Even more telling: 62% of employees feel their leaders underestimate AI’s emotional and psychological impact.

The employer brand damage is measurable. In October 2025, U.S. employers announced 153,074 job cuts, the highest October figure in more than twenty years. Year-to-date cuts reached 1.10 million, about 65% higher than 2024.

For HR and employer brand leaders, this turns layoffs into a reputational stress test. Candidates research companies long before applying. They see the headlines. They read the reviews. And they draw conclusions about which organisations view employees as assets versus costs.

What are forward-thinking companies doing instead?

The organisations positioning themselves best are not choosing between AI and people. They are investing in both.

The gap between intent and action is striking. A 2024 BCG study found that while 89% of executives say their workforce needs improved AI skills, only 6% have begun upskilling “in a meaningful way.” IBM’s Institute for Business Value estimates about 40% of the workforce needs reskilling over the next three years.

Companies that are doing it well share common approaches:

Cognizant is working to train more than one million individuals by 2026 through their Synapse program, partnering with governments, academic institutions, and businesses to create pathways in AI and digital skills.

IBM’s Skills Academy focuses specifically on AI readiness, providing structured learning paths that help existing employees transition into AI-augmented roles rather than being displaced.

SkyHive (acquired by Cornerstone OnDemand) uses AI itself to map skills at a granular level and identify reskilling pathways, helping organisations find “skill adjacencies” where employees can transition with targeted training.

How does AI replacement affect your employer brand?

Your approach to AI is now a talent attraction variable. And only 21% of employer brand teams describe their employer brand as “very mature,” which means most organisations are underprepared for this shift.

The paradox is real: the best candidates want to work with cutting-edge AI tools, but they do not want to work for companies that will replace them with those tools. They are evaluating your AI strategy as part of their employment decision.

What candidates watch for:

• Consistency: Does your careers site, investor deck, and internal town hall tell the same story about AI?
• Honesty: Are you transparent about how AI will change roles, or do you pretend nothing is happening?
• Support: When you do reduce headcount, do you provide real transition support or just severance?
• Investment: Are you funding reskilling programs, or just announcing them?

Starting August 2026, the EU AI Act classifies AI use in recruitment as “high-risk,” requiring human oversight and bias audits. Organizations hiring within the EU must ensure their AI systems meet these standards or face regulatory penalties and brand damage.

Should your company replace workers with AI?

This is not a simple yes or no. The right answer depends on your specific context, but the data suggests a framework.

The organisations that treat this as a binary choice (replace or do not replace) are missing the point. The real question is: how do you use AI to make your people more valuable, not less necessary?

Companies known for AI layoffs will struggle to attract the talent they need to execute their AI strategies. The people best equipped to work alongside AI are also the people with the most employment options. They will not choose organisations that view them as costs to be eliminated.

The companies that emerge from this transition with their talent pipelines intact will be the ones that treat workforce transformation as a long-term narrative, not a short-term cost-cutting exercise. They will invest in reskilling before replacement. They will be transparent about changes. And they will recognise that their AI strategy is now inseparable from their employer brand.

The question is not whether AI will change your workforce. It is whether you will be the company that talent runs toward or away from when that change arrives.

Frequently Asked Questions

Will AI replace most jobs by 2030?

Current projections suggest AI will significantly transform rather than eliminate most jobs. The World Economic Forum estimates 85 million jobs will be displaced by 2026, but also projects 97 million new roles will emerge. The net effect depends heavily on industry, role type, and how organisations choose to implement AI (replacement versus augmentation).

How do I know if my job is at risk from AI?

Jobs with high AI exposure typically involve routine cognitive tasks, data processing, and standardised interactions. Data entry (95% automation risk), customer service (80%), and paralegal work (80%) face the highest exposure. Roles requiring complex judgment, creative problem-solving, and relationship building have lower near-term risk.

What should companies do instead of replacing workers?

Leading organisations are investing in reskilling programs that help employees transition into AI-augmented roles. This includes mapping current skills to AI exposure, identifying “skill adjacencies” for transition, and providing structured training. Companies like Cognizant and IBM have committed to training millions of workers in AI-related skills.

How does AI layoffs affect a company’s employer brand?

Companies known for AI-related layoffs face measurable talent attraction challenges. Glassdoor data shows a 41% increase in “fatigue” mentions in company reviews, and employee concerns about AI job loss have jumped from 28% (2024) to 40% (2026). Candidates research employers before applying and factor AI strategy into their decisions.

What is Moltbook and why does it matter for the workforce?

Moltbook is a social network built by and for AI agents, where 1.5 million AI agents joined in four days without human participation. It demonstrates that AI agents can now collaborate autonomously on complex tasks. For workforce planning, it signals that knowledge work involving coordination and content creation will face increasing AI capability.