The $4.4 Trillion Question
In February 2024, something remarkable happened in Swedish fintech. Klarna, the $6.7 billion buy-now-pay-later giant, announced that its AI assistant — built on OpenAI's technology — had handled two-thirds of all customer service chats in its first month of deployment. That's 2.3 million conversations, doing the equivalent work of 700 full-time human agents.
The results were staggering: customer satisfaction scores matched those of human agents, resolution time dropped from 11 minutes to under 2 minutes, and repeat inquiries fell by 25%. CEO Sebastian Siemiatkowski projected $40 million in profit improvement for 2024 alone. By the time Klarna filed for its IPO in September 2025 — raising $1.37 billion on the NYSE — the company had reduced its headcount from roughly 5,000 to 3,422 employees, with AI shouldering an ever-growing share of operations.
Klarna isn't alone. Across industries, a new paradigm is emerging: the zero-human company — organizations that operate with minimal or no human employees, relying instead on AI agents, automated workflows, and algorithmic decision-making to deliver products and services at scale.
Vision & strategy
Execution layer
Workflows & ops
Scale without headcount
Mapping the Landscape: Who's Actually Doing This?
The zero-human company movement isn't a monolith. It spans a spectrum from "AI-augmented" organizations that have dramatically reduced headcount, to fully autonomous operations with no human employees at all. Here's the current map:
| Tier | Model | Humans | Examples |
|---|---|---|---|
| Tier 1 | AI-First Downsizers | Hundreds to Thousands | Klarna, IBM, UPS, Dropbox |
| Tier 2 | AI-Native Builders | Tens to Hundreds | Cognition, Midjourney, Harvey AI |
| Tier 3 | 1-Person Billion $ Co | 1-3 | Levelsio, Postma, Tossell |
Tier 1: AI-First Downsizers
These are established companies aggressively replacing human labor with AI:
Klarna remains the poster child. With revenue of $2.81 billion in 2024 and a workforce shrinking from ~5,000 to ~3,400, the company demonstrated that AI could handle customer-facing interactions at scale without sacrificing quality. Siemiatkowski has been vocal about his vision: a company where AI handles everything that can be measured, and humans focus exclusively on judgment calls and creative strategy.
IBM made headlines when CEO Arvind Krishna announced a hiring freeze for roles that could be replaced by AI, estimating roughly 7,800 positions — about 26,000 employees in back-office functions like HR — could be automated within five years. By 2025, IBM had begun deploying its own AI (watsonx) to handle internal HR queries, procurement, and IT support.
Tier 2: AI-Native Builders
These companies were built from the ground up to operate with minimal human involvement:
Cognition AI (Devin)
Founded November 2023 by IOI gold medalists. Built Devin, the "AI software engineer." Valued at $10B+ after acquiring Windsurf in 2025. Just 111 employees building AI that replaces thousands of developers.
Autonomous CodingJasper AI
AI content platform that generates marketing copy, blog posts, and social media content. Reached $80M ARR with under 250 employees. The content it produces would require thousands of copywriters.
Autonomous ContentHarvey AI
Legal AI platform backed by Sequoia, used by elite law firms. Automates legal research, document drafting, and due diligence. Valued at $1.5B. A small team replacing armies of junior associates.
Autonomous LegalMidjourney
Image generation platform generating $200M+ ARR with a team of roughly 40 people. No VC funding, profitable from day one. Replaced entire illustration and stock photography workflows.
Autonomous DesignCognition AI is perhaps the most radical example. Founded in August 2023 by Scott Wu, Steven Hao, and Walden Yan — all International Olympiad in Informatics gold medalists — the company released Devin, an AI system capable of autonomously completing software engineering tasks. By March 2025, it had raised funding at a $4 billion valuation from 8VC, and by July 2025, after acquiring the Windsurf IDE, it was valued at over $10 billion. All with just 111 employees.
The implications are profound: Cognition's tiny team has built a system that can, in many cases, do the work of software engineers. Devin 2.0, released in April 2025, operates as a full IDE-embedded agent that can plan, write, debug, and deploy code with minimal human intervention.
Tier 3: The "1-Person Billion-Dollar Company"
Sam Altman's prediction of the "one-person billion-dollar company" is becoming reality faster than anyone expected. Several founders are already running multi-million-dollar operations solo, with AI handling everything from customer support to product development:
The Lattice Controversy: When Is It Too Far?
Not every attempt at AI workforce integration has gone smoothly. In July 2024, Lattice, the HR platform, announced plans to create "digital workers" — AI agents that would be onboarded, managed, and tracked within their HR system just like human employees, complete with assigned managers and performance reviews.
- Tracks AI agent performance like any worker
- Unified management dashboard
- Clear accountability chains
- Formalizes what's already happening
- Normalizes human replacement
- Blurs human/machine boundaries
- Devalues human workers psychologically
- Public backlash forced full reversal
The backlash was immediate and fierce. HR professionals, labor advocates, and even Lattice's own customers recoiled at the idea of treating AI systems as equivalent to human workers. Critics argued it would normalize the replacement of humans by making AI "employees" seem natural and inevitable. Within days, Lattice reversed course, with CEO Sarah Franklin publicly walking back the initiative.
The Lattice episode revealed an important fault line: while the technology to replace workers exists, the social and organizational frameworks haven't caught up. Companies can deploy AI agents, but trying to formalize them as "digital employees" within human HR systems hit a nerve that pure automation didn't.
The Chinese Model: Speed at Scale
While Western companies debate ethics and governance, Chinese companies are moving fast. ByteDance, the parent company of TikTok, has been at the forefront of AI-driven content moderation and recommendation, using automated systems to manage billions of pieces of content with minimal human review.
| Dimension | Western Approach | Chinese Approach |
|---|---|---|
| Speed | Cautious, debate-first | Ship fast, iterate later |
| Regulation | Heavy (EU AI Act, GDPR) | Government-aligned, fast-tracked |
| Ethics Framing | Worker protection focus | Efficiency & modernization focus |
| Scale | Case-by-case deployment | Massive, population-scale rollout |
Market Size and Growth
The autonomous enterprise market is massive and growing rapidly:
Why Now? The Three Convergences
Three technological trends are converging to make zero-human companies viable for the first time:
What This Means
The zero-human company is not a thought experiment. It's happening now, across sectors, at every stage of the company lifecycle. The question is no longer if companies can operate without humans, but which functions still require them, and for how long.
At BRNZ, we're watching this space closely — and building the tools to help founders navigate it. The companies that will thrive aren't the ones that replace all humans. They're the ones that find the optimal balance between human judgment and AI execution.
The rise of zero-human companies isn't about eliminating people. It's about eliminating unnecessary work — and freeing humans to focus on what only humans can do: taste, judgment, creativity, and care.