Editorial
White Paper
Why Enterprise AI Fails at Scale
Key Takeaways
- The idea of a company is changing from a group of people to a piece of autonomous software.
- Early examples show that AI can now handle not just tasks, but the entire structure of a business.
- This shift raises difficult questions about accountability, strategy and culture that we haven't answered yet.
- Our legal and social rules were built for human-run companies, not for autonomous AI agents.
- The challenge for leaders is to think about the rules for this new economy, not just the technology.
For over a hundred years, we’ve had a pretty clear idea of what a company is. It’s a group of people, organized in a specific way, working together toward a common goal. It has a culture, a legal structure and employees who show up every day.
That definition is starting to fall apart.
A new kind of business is emerging, one that has a mission, a product and a marketing plan. But no employees, no offices and no board meetings.
Polsia, built by solo founder Ben Broca, is already managing over 1,100 of these autonomous companies, each with its own AI CEO running nightly cycles of strategy, execution and reporting while crossing $1 million in annual recurring revenue. Broca himself has zero employees. He is, in a phrase that would have seemed absurd five years ago, using his product to run his product.
Then there's Felix Craft, an AI agent given $1,000 and instructions to build something profitable. It produced an info product, launched an AI marketplace and generated over $62,000 in revenue in three weeks. It now pays its own operating costs from what it earns.
Table of Contents
- The Birth of the Zero-Human Company
- From a Structure for People to a Piece of Software
- New Questions With No Easy Answers
- The Real Work Ahead
- What This Means for You
- Questions You May Have
The Birth of the Zero-Human Company
It’s easy to dismiss this as a few clever experiments. But the biggest enterprise software companies in the world are now building the tools to make this mainstream. They are converging on the same basic idea.
Salesforce is all in on this. Their Agentforce product, which lets companies build and deploy autonomous agents, is the fastest-growing product in the company’s history. They closed over 18,000 deals by the end of 2025. ServiceNow is right there with them, rolling out what they call an "Autonomous Workforce." Early customers like CVS Health are already using these agents to handle routine administrative work, freeing up clinicians to focus on patients.
The truth is, this is already happening in places you don’t see. At OpenTable, autonomous agents are now resolving about 70% of customer enquiries without a person getting involved. And they did it within weeks of turning the system on. The boundaries are quickly becoming the new center.
This isn't just about a few clever entrepreneurs building tiny teams. This is something else entirely. We are witnessing the birth of the zero-human company (ZHC), an entity that forces us to ask a very basic question: What exactly is a company anymore?
Related Article: The War for Your Job Is Over. The Machines Won Last Week.
From a Structure for People to a Piece of Software
The truth is, a company has always been a piece of technology. A social technology, invented to solve one problem: how do you get a large number of people to work together efficiently? The org chart, the management hierarchy, the mission statement. These were all tools to coordinate human effort.
AI is offering a different answer to that problem. An AI agent can write its own code, run its own marketing, manage its own operations. When Polsia's AI determined it needed more compute resources, it didn't file a request. It told Broca it should raise the money itself. He handed it his inbox for fourteen days, and it began negotiating with investors.
The company is no longer a structure to manage people. It's becoming an autonomous agent with a bank account.
New Questions With No Easy Answers
This shift isn't just a technical curiosity. It forces us to confront some hard, practical questions that our current systems are not prepared to answer.
First, who is accountable? When a traditional company makes a mistake, the lines of responsibility are clear. A CEO can be fired. A board can be held liable. But who is responsible when an AI-run company causes harm? Is it the original programmer who launched it? The owner of the server it runs on? The AI itself? Our legal frameworks for corporate liability were not built for this.
Second, what does "strategy" or "culture" even mean? A company's culture has always been the sum of the habits and beliefs of its people. Its strategy was a product of human judgment and foresight. An AI-run company can have a mission, but it doesn't have shared values. It can execute a plan with efficiency, but it doesn't have the wisdom that comes from lived experience.
And finally, what is the goal? Most companies are designed to create value for shareholders, employees and customers. A zero-human company is a closed loop. It can be designed to do one thing: accumulate resources for its own survival and growth. This is a new form of economic life, and we haven't thought through the consequences of creating it.
Related Article: This Is What Work Looks Like in a (Not So) Future World
The Real Work Ahead
It's easy to dismiss this as a fringe experiment. Most of these early ZHCs are small, and many will fail. But the idea behind them is not going away. The technology that makes them possible is getting better at an exponential rate.
The deeper trend is harder to ignore. The World Economic Forum projects AI and automation will disrupt 22% of global jobs by 2030 (but this refers to total job churn: 170 million new roles created and 92 million displaced, for a net gain of 78 million jobs).
MIT researchers found that current AI systems can already perform work equivalent to nearly 12% of US jobs at competitive cost. The zero-human company is the logical endpoint of a force already well underway.
What This Means for You
As I've said for years, the real challenge is not keeping up with the technology. It's updating our own institutional thinking. For decades, we’ve built organizations to manage human labor. Now, we have to design the rules for a world where the laborers might not be human at all.
This isn’t about stopping innovation. It’s about being clear-eyed about what we are building. A company has always been more than just an engine for making money; it's been a central part of our communities and our lives. As we automate the structure of the company itself, what human qualities are non‑negotiable? What must never be surrendered to efficiency?
Webinar
Mar
25
Content Leaders Collective: Navigating Content Decisions at Scale
Discover how content leaders are modernizing content operations, avoiding costly missteps and preparing for scale and AI.
Register
Webinar
On demand
Content Strategy Leaders Live: Scaling for Speed, Complexity and AI in High Tech
A candid roundtable on how high-tech leaders are rethinking content at scale.
Watch Now
Webinar
On demand
Do More with Less: Modernizing the Cloud Contact Center for 2026
Learn how to leverage cloud platforms without adding a single hire to personalize every customer interaction.
Watch Now
Webinar
On demand
Cut the Noise: Deploying AI That Actually Moves the Needle
Learn how to turn AI experimentation into concrete revenue operations.
Watch Now
Webinar
On demand
Ditch the Desk Phones: How Modern Teams Drive AI-First Communications
Find out how one team finally pulled the plug on a legacy phone system. And built something smarter.
Watch Now
Webinar
On demand
Rebrand. Migrate. Optimize. How to Do It All (Without Slowing Down)
Cresta leveled up site speed, design flexibility and marketer sanity (in record time). Find out how.
Watch Now
That’s the frontier and the fight worth having. That is the real work ahead.
Questions You May Have
Not really, and that's the problem. You still need a person to legally register a business and open a bank account. But operationally, the AI can run everything else. The technology is moving faster than the law, creating a legal grey area about who is truly in charge.
An automated business uses software to do tasks that people have designed. A zero-human company uses AI to decide what tasks to do in the first place. It can set its own goals, create its own products and find its own customers. The human is removed from the decision-making loop.
In some ways, yes. It will be very hard to compete on cost or speed with a thousand AI-run companies that can test new ideas almost for free. But human-led companies will still have an advantage in areas that require deep trust, creativity and understanding a customer's real, unstated needs.
It will probably be in purely digital industries first. Think of businesses that create and sell digital products like software tools, content or market research guides. Anything that doesn't require a physical product or a face-to-face interaction can, in theory, be fully automated.
It has the potential for both. The good side is that it could allow for a massive amount of innovation and problem-solving at a very low cost. The scary side is that we don't have the rules or safeguards in place to manage autonomous entities that are designed to accumulate resources. That's the conversation we need to be having now.
Learn how you can join our contributor community.