Kristin Jin, who left Microsoft in early 2026 to start the AI implementation consultancy trnsfrmAItn, leverages AI for tasks such as proposal, pricing, and strategy, and is considering AI agents to perform some of the labor-intensive client processes.
Christine Jin
While tech giants need hundreds of billions of dollars to scale artificial intelligence, the smallest companies in the U.S. economy need just a few AI tools to get cost-competitive business models up and running.
Alex Willen expects his 12 e-commerce brands, which operate without employees, to generate $2.5 million to $3 million this year. Kristin Ginn left Microsoft in March to start an AI implementation consulting firm, trnsfrmAItn. Seven months later, with no employees or regular contractors, she received or contracted approximately $250,000 in business by the end of the year.
“The biggest misconception is that it’s just a matter of individual productivity,” said Erik Brynjolfsson, an economist at Stanford University. “In reality, it’s about leveraging AI to perform tasks that previously required entire teams.”
Early research on the impact of AI in the self-employed business world is noteworthy.
Riya Paragashvili, a labor economist and director of the Labor Policy Project at George Mason University’s Mercatus Center, says new business creation is increasing in direct correlation to applications from unlikely employers.
In industries highly exposed to AI, business applications without early signs of employment increased by 26.8% from the first quarter of 2024 to the first quarter of 2026, from approximately 222,000 to 281,000 applications per quarter.
Over the same period, applications in industries where there is a stronger indication of eventual employer potential fell by 6.4%. In construction and wholesale trade, which Paragashvili used as a comparison group, non-employer type claims fell by 0.4%.
“What AI is doing is putting these capabilities directly into the hands of individuals,” she said.
Paragashvili cautioned that the findings are preliminary and descriptive and do not prove that AI was responsible for the increase. Census-based measurements identify applications without metrics related to short-term payroll or hiring plans. There is no indication whether these companies ultimately remained privately owned or employed no employees.
But Willen shows us what the economy is like.
Anthropic’s Claude helps with bookkeeping, customer service, advertising, and inventory management across the business, and OpenAI image tools perform tasks once done by photographers and designers.
Without AI, Willen said, “We would probably need another full-time person, and we would need to use contractors for photography, design, bookkeeping, and advertising management.”
He said traditional product photography can cost $2,000 to $5,000, and designers can charge $300 to $1,000 to shoot new listing images.
If AI directly reduces one of these expenses, “that money just goes into my pocket,” Willen says.
Thanks to AI, you can now pursue revenue channels you never considered before. After acquiring Lone Trail, a hunting equipment brand that sells binocular harnesses, packs and other outdoor accessories, Willen said the company’s Shopify site accounted for about 1% of its sales. Rather than shut down the channel, he rebuilt the site and AI using ChatGPT and Claude to help run meta ads. He said Shopify accounted for about 30% of Lone Trail’s revenue in the last 30 days.
“The end result is at least six-figure annual revenue that wouldn’t exist without AI,” Willen said.
Ginn’s business shows how quickly new entrants can reach meaningful revenue without first building an organization. She uses AI to help with recommendations, pricing, strategy, and other tasks, and is considering AI agents to perform some of the labor-intensive client processes.
A study conducted by accounting and invoicing software company FreshBooks identified similar practices more broadly. 79% of self-employed individuals said they had tried AI before hiring someone for tasks they couldn’t do themselves (the same number was 86% when including small business owners). Shaheen Javadizadeh, CEO of Freshbooks, says AI makes books “seem much bigger than they really are.”
However, for established small businesses, the available data tells a different story. Gusto, a payroll and HR software company serving small and medium-sized businesses, tracked companies that adopted AI and found that they had about 7% more employees one year later than comparable companies that knew about AI but didn’t implement it. For companies with fewer than 10 employees, companies that adopted AI increased their headcount by approximately 10% compared to comparable companies that did not.
Willen’s experience also highlights the tradeoffs in new business models. In other words, operating with fewer people does not preclude dependence on large technology platforms. About 80% of his overall revenue still comes from Amazon, and he uses Meta and Shopify to grow his direct sales. Although Jin doesn’t have a staff, she said LinkedIn is one of the biggest factors in attracting clients to her consulting firm.
Brynjolfsson said the proliferation of small producers could actually increase the value of platforms that connect them with buyers.
“There’s a long tail of products, but at the same time, big platforms like Amazon are more powerful than ever by aggregating all of these things and bringing them together,” Brynjolfsson said.
However, this relationship is not so simple that labor savings flow directly to Big Tech. Willen said Ginn spends less than $60 a month on paid AI tools, although some of the money he saves goes to profit.
AI will enable more small-scale producers to exist economically. And as core business production has become cheaper, small businesses have become more focused on customer acquisition and sales, Brynjolfsson said.
Adam Branch, who founded rhino.ai, an enterprise software company that uses AI to modernize legacy systems, previously founded a 500-employee technology company. Now, he says, AI software “will do a lot of the work that teams used to do.”
adam branch
Brynjolfsson believes that while individual entrepreneurs will flourish, the use of AI will result in the creation of even larger companies on average than today.
Brynjolfsson and Zoe Hitzig, a researcher at the Institute of Anthropology who studies the social and economic impacts of AI, argue in a September research report titled “The Use of AI Knowledge in Society” that by codifying knowledge that was previously scattered among workers and enabling them to aggregate more information and act centrally, companies will become larger on average, while industry concentration will increase and local management autonomy will decline. Researchers say early evidence is broadly consistent with these patterns.
This is bad news for midsize companies. Brynjolfsson predicts that AI could contribute to the “hollowing out” of midsize companies. This means that very small teams can do more on the one hand, and huge organizations leverage AI and scale on the other.
“The fat in the middle, the management layer focused on internal coordination, is likely to be significantly reduced,” he said.
Adam Branch experienced that change within two very different technology companies.
Before founding rhino.ai, an enterprise software company that uses AI to modernize legacy systems, Branch founded Incentive Technology Group, a traditional technology services company that grew to approximately 500 employees. Under that model, he said, getting more jobs means hiring more people.
“With rhino.ai, we are building AI software that does much of the work that these teams were previously doing,” Branch said.
Rhino.ai has approximately 50 employees. We estimate that branches can perform equivalent modernization efforts at three to four times the productivity per employee.
His team estimates that roughly 85% of ITG’s operating expenses are labor costs, with technology accounting for about 5%. At rhino.ai, Branch says the workforce is about 40%, while technology, including cloud, software and AI models, is about 25%.
AWS sees a similar moderation in the relationship between revenue and headcount among AI-native startups. The company’s Growth Engine Study, an independent global survey of 3,413 startup founders and senior leaders in 20 countries, found that these companies generated an average annual revenue increase of 156%, compared to 65% for startups overall, with more than half generating more than $400,000 in revenue per employee. This report focuses on companies less than five years old that are using AI in cutting-edge forms.
“Growth used to be directly related to the number of people you could hire,” said Luba Borno, vice president of AWS Specialists and Partners. “Founders can now scale not just in number of employees but also in technology.”
Brynjolfsson doesn’t see this result simply as a victory for Big Tech at the expense of everyone else. Entrepreneurs will be able to keep some of the productivity gains, consumers will have more choice, and platforms will benefit by providing technology and connecting more producers with customers, he said.
“People who are creating platforms and tools are winning because they can sell the tools, independent entrepreneurs are winning because they can use the tools, and consumers are winning because they can get new products that weren’t available before,” he said. “It’s mainly a bigger pie, so it’s a win-win-win, a triple win.”
What AI could change, he says, is the shape of a larger pie.
“It’s clear that every time there is a change in technology or a change in platforms, there is a benefit for the companies that are developing those platforms and technologies,” Brynjolfsson said. “But there are also benefits for newcomers who can start fresh.”
