One-person company

(Image caption) August 8, 2026, Social Mixer event celebrating the first anniversary of Solo Unicorn Club, New York.

An anniversary event without a stage

On August 8th, Solo Unicorn Club held its first anniversary Social Mixer event in New York. This event featured no keynote speeches, no roundtable discussions, and no entrepreneurs lining up to present their projects. 103 applications were approved, with 48 participants having attended previous events. The attendees included entrepreneurs, AI engineers, independent developers, investors, as well as business leaders and professionals from the finance, building technology, and life sciences industries.

For a city like New York, which is never short of tech events, investment forums, and startup gatherings, an event without a stage is nothing to brag about. What Solo Unicorn Club really wants to do is quite practical: put people who have the ability to build products next to people who truly understand business issues, and dedicate time to conversations that may lead to future collaborations.

An independent developer may have already created a product prototype using AI, but he doesn't know where his first enterprise customer is; a business manager may know exactly which processes in their company are inefficient or which problems have remained unresolved for years, but he may not know that a group of people have emerged in the market who can quickly test new solutions with very small teams. Investors are not just concerned with who has the latest model, but whether these technologies can ultimately find stable demand, generate revenue, and build sustainable companies.

Solo Unicorn Club navigated its first year amidst these specific and less-than-glamorous issues.

Founded in August 2025 and based in New York, this AI builder community currently boasts over 2,000 online and offline members. One reason for its initial attention was its advocacy and support for the emerging "one-person company" model, while simultaneously incorporating AI-native transformation within enterprises into a unified community framework. These two seemingly disparate entities—small startups and large corporations with complex organizational structures—both face the same transformation: AI is enhancing individual capabilities and changing how companies allocate human resources, technology, and capital.

(Image caption) Dr. Jesse Qin, founder of Solo Unicorn Club.

A "one-person company" does not mean that one person does everything in isolation.

"One-person company" is a concept that is easy to spread and can be easily misunderstood as meaning that future companies will only need one employee, or even further interpreted as "one person plus AI can eliminate the need for a team".

Dr. Jesse Qin, founder of Solo Unicorn Club, does not see it that way.

In his view, what AI truly changes is the previously relatively fixed relationship between company size and production capacity. In the past, if a company wanted to increase its products, expand its market, and serve more customers, it usually needed to simultaneously increase its engineering, design, sales, research, content, and operations staff; now, some of these tasks can be assisted by AI, and an experienced builder or a small team can handle tasks that previously required more manpower.

"I don't believe that a one-person company means one person replacing all the functions of a company. The real change is that the minimum effective size of a company is decreasing," said Jesse Qin. "In the future, when evaluating a company, the number of employees may no longer be as representative of its actual capabilities as it used to be. We need to look more at what capabilities it has mastered, what problems it has solved, and to what extent AI has amplified the capabilities of each individual."

This is a more worthwhile question to discuss than "Can one person create a billion-dollar company?"

Today, an entrepreneur can use large models to conduct research, leverage AI to assist in programming, quickly build product prototypes, and compress aspects of design and operations that previously required multiple people into a small team. This reduces the human and time costs in the early stages of a startup, meaning an idea no longer needs to raise substantial funding or assemble a full team before it can be tested in the market.

This capability has brought opportunities to many that were previously unavailable, but it hasn't simplified business itself. Products can be made faster, but customers don't automatically appear; the cost of code has decreased, but corporate procurement, legal liabilities, data security, regulatory requirements, and customer trust still exist. Technology can help a company become lighter, but it cannot take away the responsibilities a business should bear.

Jesse sees this line very clearly. "A company is not just a collection of production tools; it also involves trust, legal responsibility, customer relationships, governance, and long-term delivery. AI can reduce some production costs, but it cannot create these things automatically. Even if a company has only one person, contracts still need to be fulfilled, customers still need to be served, and responsibilities still need to be borne by someone."

This is perhaps the most important premise for understanding "one-person companies." AI making a person more capable does not mean that the business world no longer needs other people.

As products are developed faster, the truly valuable questions become even more precious.

Several discussions that took place at the first anniversary event clearly illustrate the true state of AI in enterprises.

An AI executive at a major investment bank discussed the deployment of intelligent document systems within the company. The real challenges extend beyond the models themselves; they include how to build the knowledge graph, handle token restrictions, determine which models are allowed under existing contracts, explore the viability of open-source alternatives, and ultimately, who should approve the system and which department should bear the costs.

A product manager at a global building technology company was discussing the AI-driven transformation of HVAC and cooling systems. AI technology can be updated very quickly, but a commercial building may have been operating for twenty years, with existing equipment, control systems, maintenance responsibilities, procurement procedures, and decision-making processes between different departments that cannot be completely rebuilt just because a new model emerges.

The situation is even more typical in the life sciences. While gene sequencing, cell therapy, and cancer research do generate massive amounts of data and there is room to leverage AI to improve efficiency, research validation, patient safety, clinical responsibility, and regulatory requirements still have their own pace. AI can shorten certain steps, but it cannot replace the responsibilities that science and medicine must bear with speed.

These seemingly unremarkable details are precisely what make AI the most challenging aspect of truly entering the commercial world.

A few years ago, one of the biggest obstacles for an entrepreneur might have been the lack of a technical team—knowing a problem but lacking the ability to build a product. Today, that's changing. Large models, agents, cloud infrastructure, and a wealth of open-source tools have made engineering capabilities more readily available than ever before; a good builder can create a fairly complete prototype in days or weeks. With this increased speed of product development, another scarcity is emerging: what problems are truly worth solving?

Jesse believes that many excellent builders today lack not technical skills, but rather on-site industry knowledge. "Who truly has this problem? How painful is it? Does the company have the budget? Who has the purchasing power? Will solving it generate measurable value? These questions may not sound as exciting as model parameters, but they ultimately determine whether a product can become a company."

This assessment also explains why Solo Unicorn is willing to invest effort in bringing actual business operators into a community primarily focused on AI builders. The tech side needs to understand how real businesses operate, and the businesses need to see what small teams are already capable of. Neither side lacks capability; what they lack is a channel that truly allows needs and capabilities to meet.

(Image caption) Before the event, the community helped most of the confirmed attendees with background information and arranged icebreaker groups, with personalized name tags becoming the first entry point for conversation.

The most important knowledge of a company is often not fully written in documents.

Jesse Qin's emphasis on "context" is directly related to his professional experience.

He holds a Ph.D. in Computer Engineering from Rutgers University, with research interests including distributed computing, recommendation and knowledge discovery, and machine-readable knowledge systems. In 2025, he completed his MSBAi in Business Analytics and Artificial Intelligence at NYU Stern School of Business. Following this, he worked at Kamiwaza, an enterprise AI infrastructure company, on product engineering related to ontology, context graph, RAG, and agent workflow. His public research on Context Graph also focuses on enterprise knowledge fragmentation and how humans and AI agents can share governable organizational context. In May 2026, he publicly shared his research at the Knowledge Graph Conference 2026 at Cornell Tech in New York.

Therefore, his statement that "the ceiling of a one-person company may not be production capacity, but the context" is not a slogan designed for the entrepreneurial community, but a judgment he formed after long-term research on enterprise AI.

Jesse explained to GFM.News that many people tend to believe that as long as the model's capabilities continue to improve, someone with sufficiently good AI tools can handle more and more tasks. From a productivity perspective, this trend has indeed emerged, but in actual business operations, a large amount of important information does not exist in publicly available data, databases, or documents that can be directly handed over to AI for reading.

Why is a bank particularly cautious at a certain stage? Why must a pharmaceutical company insist on a certain validation procedure? Why did a company abandon a seemingly reasonable product direction many years ago? There may be regulatory reasons, or it may stem from a costly failure, a request from an important client, or a collective judgment gradually formed by a group of employees who have worked there for many years.

“A model can read a lot of data, but it doesn’t automatically acquire the experience an organization has accumulated over ten or twenty years. That experience is part of the context,” Jesse said. “A builder can quickly make a product today, but he may still not know what the company really needs, so I’ve always believed that the people in the company who have actually done the work are a valuable source of knowledge.”

Here's a change that businesses and entrepreneurs should pay attention to. As AI becomes more powerful, the value of industry experience is not declining as rapidly as some might imagine; in fact, it may become even more prominent due to lower engineering costs.

A skilled developer might be able to build a beautiful prototype in three days, but a manager with twenty years of experience in the same industry could explain in a half-hour conversation why the product wouldn't be adopted in a real enterprise environment. For an early-stage startup, that half-hour could save not three days, but six months.

Solo Unicorn aims to establish the conditions necessary for this kind of communication to occur.

Not everyone needs to be an engineer, nor do all engineers need to spend twenty years relearning an industry. AI's efficiency can only translate into real business efficiency if industry experts are willing to explain the problems clearly, capable individuals are willing to solve them, and companies can provide real-world environments for validation.

(Image caption) Independent developers and enterprise managers exchange real business issues on site.

A company's capabilities may become increasingly difficult to measure in terms of the number of employees in the future.

The issues discussed by Solo Unicorn deserve to be viewed within a larger economic context, and for this very reason.

For a long time, modern enterprises have followed a relatively stable pattern: business expansion is often accompanied by an increase in employees. More product lines require more engineers; more customers require an expanded sales and service team; and as the organization grows larger, it necessitates more management layers and administrative costs. Thus, enterprise output, employee numbers, and capital investment maintain a relatively close relationship.

AI is gradually loosening this relationship.

If a five-person company can do the work that used to be done by twenty people, and a twenty-person enterprise can support the business that used to require one hundred people, then our understanding of corporate efficiency and corporate value will inevitably change. The number of employees is still important, but it can no longer directly represent a company's actual productive capacity as it once did.

For investors, future priorities may include how much revenue each employee can generate, how many of the company's processes have truly been redesigned by AI, and whether revenue growth still requires a proportional increase in staff. For managers, the questions will gradually shift to another level: which tasks are best suited for AI to improve efficiency, and which processes must still retain human experience, judgment, and responsibility.

The impact of this change on the labor market is far more complex than simply whether a particular job will be replaced by AI.

Many professions may not disappear in the short term, but the boundaries of an individual's work may expand significantly. Some tasks that were previously scattered across several positions may be recombined into one person's job; the scope of responsibility that an excellent employee can assume with the assistance of AI may also be much larger than before.

As a result, companies will rethink the value of talent. The truly scarce people in the future may not just be those who can complete a standardized task, but rather those who can ask the right questions, understand complex contexts, make judgments, and are willing to take responsibility for the results.

AI can increase human productivity, but it is difficult to replace a person's judgment formed over many years.

This is an issue that education, vocational training, and corporate talent development will inevitably have to address sooner or later.

AI can reduce the cost of ordinary collaborations and potentially increase the value of high-quality relationships.

When discussing one-person companies, Jesse particularly emphasized another often overlooked issue: just because one person has greater productivity does not mean that he no longer needs others.

He even believes that the situation may be exactly the opposite.

"The biggest misconception about one-person companies is that they mean one person doesn't need anyone else. I don't believe in that kind of business world," Jesse said. "AI can enable one person to complete more production tasks independently, but business itself is still a form of social collaboration. You still need customers, partners, industry knowledge, legal, financial, and various professional services. More importantly, you need others to trust you."

This is a very humane and practical reminder.

A small company may no longer need ten full-time employees, but it still needs a trustworthy lawyer, an advisor who truly understands the industry, a client willing to provide the first use case, and a few partners who can tell the truth in critical moments.

Previously, companies relied on a large number of internal staff to complete collaborations; in the future, some smaller AI-native companies may use fewer full-time employees, coupled with higher-quality external professional networks, to accomplish the same or even larger tasks. This does not mean that interpersonal relationships become worthless, but rather that the quality of those relationships becomes more important.

AI can handle a large amount of routine information exchange for humans and reduce some collaboration costs, but trust is difficult to generate automatically. Especially in the business world, whether an individual is willing to introduce their clients to you, whether a company is willing to let your product enter its core processes, and whether a seasoned industry professional is willing to share their years of accumulated experience with you, all ultimately depend on human judgment.

If Solo Unicorn ultimately generates long-term value, it may not lie in how many people it brings together in one space, but rather in whether it increases the probability of those who truly need each other meeting.

(Image caption) The first anniversary event concludes. Solo Unicorn Club plans to expand its community network to Los Angeles, Silicon Valley, and Tokyo.

Clubs allow people to get acquainted; Labs must prove whether these relationships can produce results.

Solo Unicorn's next phase of development will continue along the two lines of Club and Lab.

Club continues its community and connectivity efforts, and plans to gradually expand beyond New York to Los Angeles, Silicon Valley, and Tokyo. These cities possess vastly different industrial landscapes: Silicon Valley's strengths lie in its high density of technology, startups, and venture capital; Los Angeles connects entertainment, content, and the creator economy, and also boasts aerospace, healthcare, and a large number of SMEs; Tokyo, on the other hand, possesses a large and mature corporate ecosystem, while facing long-term pressures from corporate modernization, demographic shifts, and the adoption of AI.

For this type of network, adding more cities is not particularly meaningful; what truly matters is the effective flow of needs and capabilities between different regions. Whether a New York builder can access the practical problems of a Tokyo company, or whether a Los Angeles company can find a technology team not originally on its supplier list—these are the real values of such a network.

Lab, on the other hand, needs to complete the more difficult part.

Solo Unicorn currently maintains the Solo Unicorn Toolbox, which includes over 600 open-source projects related to AI infrastructure and agents, and provides bilingual (Chinese and English) materials. According to the current design, the Lab will also take on enterprise needs arising from the Club, organize these issues into specific projects, and then organize builders and industry experts to complete the Proof of Concept (POC), performance verification, and actual delivery.

Jesse knew very well that this step was much more difficult than organizing an event.

"The biggest problem with many communities is not that they don't meet a lot of people, but that once people get to know each other, the matter ends there," he said. "What I really hope to see is whether a builder gets their first client because they met a company here, whether a problem raised by a company eventually becomes a proof of concept (POC), whether the POC actually enters the business, or whether two people who didn't know each other before can eventually start a company together because of a single exchange."

Therefore, he is unwilling to measure the success of Solo Unicorn solely by the number of members.

Two thousand members is a starting point, and a number easily visible to the outside world. But for a community that aspires to build an innovation network, more meaningful numbers are often less likely to appear quickly in news headlines: how many companies are willing to share their real problems, how many problems are transformed into testable projects, how many proof-of-concepts are ultimately used, how many entrepreneurs find their first paying customers, and how many people met by chance at events eventually form sustainable business relationships.

These results may take months or even years to become clear.

Only then will Solo Unicorn truly know whether it has built a lively community or an innovative mechanism with long-term value.

What truly remains from an event should not just be photographs.

On the evening of August 8, participants gradually left the event. Personalized name tags were removed, new contacts were entered into their phones, and the next day, everyone would return to their own code, clients, companies, labs, and investment projects.

Anyone who frequently attends tech events is familiar with this scenario. Many conversations quickly fade away after the event ends, and many exciting dialogues at the time end up as a business card or a LinkedIn link.

Therefore, whether an event is valuable or not is often not something that can be judged on the day itself.

Perhaps a few months later, a developer who didn't have any enterprise clients that day will get his first contract because of a casual conversation; perhaps a process issue that a corporate manager casually mentioned will eventually be turned into a real product by another person; or perhaps two people who met for the first time that night will start a company together after six months of working together.

If these things really happen, the one-year anniversary party on August 8th, which has no stage, will have a longer life than the event itself.

In today's rapidly developing AI landscape, it's easy to focus on how much the model has improved, how much work the agent can accomplish, and how much manpower the next generation of technology will save. These are all important questions, but if we truly look at the business world, we'll find that companies exist not simply because one person doesn't have enough time to do all the work.

A company is made up of people, and it also needs the experience of others, the mutual correction of different professions, the long-term trust built by customers, the reminders from someone when problems are not seen, and a system that can ensure the continuous existence of responsibility.

AI can significantly shorten the time required to complete many tasks, but it's difficult to compress twenty years of experience needed to understand an industry into twenty minutes; it can help a person read more materials, but it cannot guarantee that they will truly understand another person's needs as a result; it can improve efficiency, but the most difficult things to build in business—credit, responsibility, judgment, and trust—still need to be built up slowly by people over time.

This is perhaps the most noteworthy aspect of Solo Unicorn Club after its first year.

Jesse Qin doesn't interpret "one-person company" as someone creating a business in isolation from the world. In his view, AI empowers individuals, so people should be more aware of what they don't know and cherish those who truly understand the industry, their customers, are willing to share their experience, and can build trust.

If companies can indeed become smaller in the future, then we may need to rethink more than just the size of companies.

We also need to rethink how a person's value is measured, where a company's capabilities come from, why industry knowledge remains important, and in an era of increasingly powerful technology, what is the true value of experiences and trust between people that cannot be quickly replicated?

For Solo Unicorn Club, the first year has come to an end.

The real test is just beginning.

The next statistic worth reporting should not be just how many members have increased, but how many people have actually accomplished something because of this network.

Disclaimer

The information in this article, including community size, participation in activities, project planning, and the founders' professional backgrounds, is based on data provided by Solo Unicorn Club. The impact of AI on company size, talent structure, and business organizations is an editorial analysis by GFM.News authors based on relevant materials.