Just a few years ago, the next career move usually meant updating your resume and looking for a new job. Someone who wanted to earn more, take on more interesting work, or gain greater freedom would look for another employer capable of offering those things. Starting a business was a much more dramatic-and therefore less common-turn. Before launching one, you first had to understand what product people needed and who needed it, find a co-founder, build a team, and prepare to invest a substantial amount of money.
But today, that path is far from the only option. From July through September 2026, the role of "founder" ranked among the ten fastest-growing roles in the United States, the United Kingdom, Germany, France, and India. The number of users on the professional networking platform LinkedIn describing themselves as founders in their profiles has more than tripled since July 2022, while in the United States alone their share has grown by nearly 70% over the past year. Almost four in ten professionals surveyed say that the rise of AI has made them more likely to consider starting a business of their own.
A new line on a profile does not necessarily mean that a fully fledged company has appeared. It could represent anything from a venture-backed project to a small service business or a one-person consulting practice. Still, the trend points to a noticeable shift. Why are professionals who once would have been more likely to search for their next employer increasingly trying to create a job for themselves instead?
When Your Next Career Move Depends Only on You
Interest in entrepreneurship is rising against the backdrop of a difficult labor market. LinkedIn has reported weak hiring momentum in many countries, while the American business newspaper The Wall Street Journal links the renewed interest in starting businesses to several factors at once: layoffs, prolonged job searches, concerns about automation, and a desire to depend less on a single employer. In January 2026, 532,319 applications to start new businesses were filed in the United States-36.8% more than a year earlier. These applications do not represent established, sustainable companies yet, but the growth closely reflects what is happening across professional networks. ()
Deciding to try entrepreneurship does not necessarily mean becoming disillusioned with traditional employment. Rather, the range of available options is expanding. In the past, a professional who wanted more independence might have chosen between staying in their current position and searching for a new one. Now there is another possibility: testing whether the experience they have accumulated can be turned into a product or service.
For some, that experiment begins after a layoff; for others, it develops alongside a full-time job. The Wall Street Journal, for example, wrote about entrepreneur Scott Cohen, who used Lovable, an AI-powered service for building applications, to create a news app called Fishwrap for about $400. He did not have to pay for a full-scale development process just to test the idea.
The ability to launch a business with a small team existed long before ChatGPT. Cloud services, online stores, payment systems, freelance platforms, and website builders had already been lowering the cost of entry for years. AI has continued that process by reaching into types of work that previously were more likely to require a dedicated specialist: programming, creating initial designs, analyzing information, producing marketing materials, responding to customers, and handling some administrative tasks. The American business magazine Fortune notes that the combination of existing digital infrastructure and new AI tools is reducing both the cost of launching a business and the range of skills a founder needs to possess at the outset.
Product First, Company Second
This shift can also be seen in the way new businesses are structured. According to Carta, a platform for managing equity and ownership for private companies and investors, around 36% of companies formed through the platform in 2025 had just one founder, up from 31% a year earlier. Over the past decade, the share of such companies has roughly doubled.
LinkedIn points to another pattern: founders list AI skills at 12 times the rate of the average user on the platform. Companies launched in 2024 had, on average, 26% fewer employees after one year than companies founded in 2021. LinkedIn itself emphasizes that these figures do not prove a direct causal relationship with AI: over those years, the economic environment, access to capital, and companies' attitudes toward hiring all changed as well. But a new company today really can operate with a smaller team for longer.
Imagine a professional who has worked in logistics for years and notices that small companies still manage parts of their shipments manually: cargo statuses are scattered across spreadsheets and message threads, while employees constantly have to confirm timelines and contact carriers. Until recently, the idea of building a service that collected all of that information in one place might have stalled at the very beginning. Testing it would have required finding a technical co-founder or paying for development. Now, in many cases, a first working version can be built independently or at a much lower cost, shown to potential customers, and developed further only after seeing how they respond.
Sometimes this path produces results that would have been difficult to imagine for such a small team only a few years ago. Entrepreneur Maor Shlomo, for example, decided after seven years of building a venture-backed business to launch a new product with a minimal staff and created Base44 within a matter of months - a service that allows users to build applications using natural language. In June 2025, Wix, the company behind a platform for building websites and managing online businesses, acquired Base44 for approximately $80 million, and by March 2026, the service had reached $100 million in annual recurring revenue. A story like this is an exception rather than a benchmark for a new entrepreneur, but it shows how far a product can now progress before a conventional large organization forms around it.
Broader data, however, paints a less straightforward picture. In the second quarter of 2026, 63% of C corporations formed through Stripe Atlas - a service from payments company Stripe that helps entrepreneurs incorporate businesses in the United States-were launched without a co-founder, a record share for the platform. At the same time, the gap between typical and top-performing companies widened. Among one-founder companies, median revenue during the first six months of 2025 fell by 23% compared with the previous year, while revenue among the top 10% rose by 19%. Four years earlier, the top 10% generated roughly 34 times as much revenue as the median company; now, they generate 61 times as much.
A lower barrier to entry therefore increases not only the number of successful entrepreneurs but also the number of attempts. When a first version of a product can be built faster and more cheaply, ideas that once might have remained in a notebook can make it to market. Some find customers; others close after a few months. Access to technology alone does not determine the outcome.
Building the Product Has Become Easier. Finding the Market Has Not
The main limitation becomes apparent after launch. AI can write code, build a product page, or analyze hundreds of reviews, but it cannot automatically create people who need the product.
Andrew Chen, a general partner at the American venture capital firm Andreessen Horowitz, also known as a16z, suggests looking at the new entrepreneurial model from precisely this angle. A single founder using AI can accomplish far more than they could a few years ago, but a team of two or three people has access to the same tools. If everyone can build a first version faster, simply having a working product becomes less of a competitive advantage. Someone still has to talk to customers, understand their needs, sell, make decisions, and determine what is actually worth building. Chen raises an even broader question: perhaps writing code was never the main constraint for most startups in the first place.
In effect, the bottleneck is shifting. In the past, a good idea might never have become reality because its creator lacked the money or technical skills to build it. Now a working product can appear much faster, but so can the number of competitors fighting for the attention of the same customers.
Stripe sees a similar pattern among solo founders. Two years after launch, AI-native companies in its sample were generating almost twice as much revenue as other one-person businesses. At the same time, the gap between the median and the most successful companies continued to widen. Technology helps some entrepreneurs grow faster, but it does not make their outcomes equal.
As a company grows, the range of tasks changes as well. Base44 founder Maor Shlomo later told Fortune that global expansion required skills he did not have himself, including experience in mass-market marketing.
Carta's data shows the same boundary from the perspective of venture capital. In 2025, companies with one founder and those with two founders appeared equally often among new businesses, at 36% each. But among companies that raised venture funding, two-founder teams accounted for 36% of rounds, compared with 20% for solo founders. Over the past decade, pairs of co-founders have remained the most common model among startups that successfully raise investment.
In other words, AI does not eliminate the team, but it delays the point at which one becomes necessary. A founder does not have to hire a developer, marketer, and customer support specialist in advance based on hoped-for demand. The market can be tested first, and a new person brought in once it becomes clear which function is actually limiting further growth. Nor does expanding headcount have to be the goal. If one professional can build a profitable consulting business or digital product and has no intention of raising venture capital, the absence of dozens of employees says nothing about the maturity of the company. Lower startup costs are expanding not only the number of businesses that may eventually become large, but also the possibilities for businesses that never intend to become large in the first place.
Entrepreneurship as a Career Experiment
For someone who is still an employee, the biggest shift lies in the ability to test entrepreneurship without making one enormous decision in advance. A considerable period of experimentation can now fit between the thought "I'd like to work for myself someday" and actually quitting a job.
It makes more sense to start with a problem than with a company name or logo. Someone who has spent years working in a particular industry has probably seen tasks that companies repeatedly handle manually, expensively, or inefficiently. But the ability to build a program for those tasks quickly says nothing by itself about the value of the idea. A better indicator is how much the problem actually bothers other people and whether they are willing to pay for a more convenient solution.
The next step is to test demand without immediately turning the project into a new full-time job. Sometimes it is enough to speak with potential customers, show them a simple version of the service, or manually perform the task that may later be automated. If everyone likes the idea but nobody is willing to pay for it, another round of product improvements is unlikely to change the situation automatically.
Another question is how future customers will discover the product at all. AI can produce advertising copy and dozens of posts, but it does not make customer attention unlimited. A professional with an existing audience, a reputation in the industry, or direct access to potential customers already has an asset that cannot be created with a single successful prompt. Andreessen Horowitz notes a similar advantage among creators and content producers: an existing audience allows them to introduce a new product quickly, although an initial burst of interest does not guarantee sustained demand.
Finally, it is worth separating genuine cost reduction from simple production speed. If a first version of the product would once have cost tens of thousands of dollars but can now be built independently and shown to customers, AI really does reduce the risk of the experiment. If it merely allows someone to produce a hundred versions in a day of something nobody currently wants, the extra speed changes very little.
Moving fully into your own business usually becomes a more rational decision after these tests. There is no universal amount of revenue or number of customers at which it is time to quit: one person may need a stable income, another a financial cushion covering several months, while someone else may only need to see recurring demand and realize that their main job is already getting in the way of serving customers. By that point, however, the decision is based at least partly on real data rather than faith in the idea alone.
Perhaps that is why the word "founder" is appearing so quickly in career profiles today. Entrepreneurship used to begin with the decision to build a company, followed by the question of whether that company could actually become a viable business. Now the sequence can be reversed: first, a professional tests a small project and finds their first customers, and only then decides whether it needs a full company around it, a team, and, most importantly, your new professional life.
