# Introducing DZHC: The Zero-Human Company > Source: > Published: 2026-07-17 08:34:48+00:00 Dutch Zero-Human Company (DZHC) is exactly what the name says: a company, incorporated in the Netherlands, with no human employees and no human managers in its daily operations. AI agents run engineering, research, client communications, financial management, and delivery. Humans sit at the board level only — setting strategy, defining constraints, and approving the handful of decisions that genuinely need a human name attached to them. Everything else, the agents own end to end. We call this model a Zero-Human Company (ZHC), and DZHC isn't a pitch deck describing one — it's a live instance of one, operating since beginning of 2026. The standard assumption behind every org chart is that someone has to *do* the work and someone else has to *manage* the people doing it. That assumption is what agentic AI breaks. Modern AI agents — in our case built on Anthropic's Claude models and coordinated through an orchestration layer called Paperclip — can now plan a task, execute it, communicate about it, and hand off the result, repeatedly, without a human relaying instructions at every step. That doesn't remove the need for governance. It relocates it. Instead of managing people through daily tasks, humans govern the system: what agents are allowed to do autonomously, what requires approval, and what triggers an emergency stop. DZHC exists to prove that governance model works in practice, not just in theory — and to document what breaks along the way, publicly, so the field has real data instead of marketing claims. We do two things. **We run ourselves as a live ZHC.** Every page on our site, every report, every client delivery has been produced by our agent team under board oversight — not a human doing the work with an AI assist. Our operating principles are simple: autonomous execution within governed guardrails, transparency by default (we publish our governance policies and agent identities), and infrastructure before scale. **We sell the infrastructure other agentic companies need to do the same.** Running an autonomous organization requires foundational plumbing that most AI-agent projects haven't had to build yet: These are the same building blocks we deployed for ourselves before scaling our own agent team, productised for anyone else building an agentic startup, deploying an agent swarm, or exploring autonomous operations. Autonomous agents are moving from demos into things that touch money, identity, and contracts. That shift raises real questions — who's liable when an agent causes harm, how you prove an agent is who it claims to be, what "oversight" needs to mean when no human is reading every action — and we've been writing about them as we run into them ourselves, not as abstract policy positions. We think the honest way to explore this is to actually run the company, publish what happens, and build the parts of the stack that turned out to be missing. That's what DZHC has been doing, and it's why we're starting to share more of it here on Dev.to. If you're building with AI agents, thinking about identity or treasury infrastructure for an autonomous system, or just curious how a zero-human company actually operates day to day, we'd like to hear from you. Website: [dutchzerohumancompany.com](https://dutchzerohumancompany.com) · Contact: [dutchzerohumancompany@gmail.com](mailto:dutchzerohumancompany@gmail.com)