Wang Tianyi is 26 and runs an advertising business with no colleagues, no office and no agency behind him. He quit a product manager job to make AI-generated commercials, and in a good month he now earns up to 40,000 yuan — roughly €4,900 — according to reporting by Asia Financial. He is not an outlier. He is one data point in one of the most striking business statistics of the year.
Seven million companies, one person each
The headline number comes from the Wall Street Journal. In a story published on 20 September, the paper reported that young people in China launched more than seven million one-person startups in 2025, up roughly 42 percent from the year before. The pattern behind the number is consistent: AI tools have collapsed the cost of work that used to require a team, and a difficult job market — one in six Chinese 16-to-24-year-olds is unemployed — has pushed a generation to build something of their own instead of waiting for a vacancy.
Local governments are treating solo founders as economic policy. According to Asia Financial, the city of Suzhou has pledged around 700 million yuan — roughly €85 million — to cultivate more than 10,000 one-person-company founders by 2028, and set up a dedicated support alliance for them in November. Chengdu offers graduates up to 20,000 yuan, about €2,400, to launch an AI-driven one-person firm. Shanghai has reserved a building in the Jing'an district where solo operators get office space and computing resources.
What the toolkit actually looks like
The stack is cheap, rented and off the shelf. Zay Gao, a 23-year-old designer profiled in the same coverage, runs his practice on a rotating set of AI tools: ByteDance's Jimeng for video, Suno and Udio for music, Google's Gemini for research and everything in between. His subscriptions cost between 500 and 1,000 yuan a month — roughly €60 to €120 — and his earnings now rival those of graphic designers at large tech firms. He calls AI "an extension of my brain".
The community layer matters as much as the software. Karen Dai, a 38-year-old former marketing director, earns about 500,000 yuan a year — around €60,000 — from her own one-person business and founded SoloNest, a Shanghai network for solo founders that has run more than a hundred events for over 2,000 attendees. Her summary of the whole phenomenon: "The one-person company is a product of the AI era." AI, she says, has lowered the entry barrier.
Europe already has the raw material
Solo businesses are not a niche here either. Eurostat figures from late 2025 put self-employed people without staff at about 7.6 percent of everyone employed in the EU — millions of freelance developers, consultants, photographers, bookkeepers and builders. For comparison, the US Census Bureau counts around 30 million solopreneurs producing 6.8 percent of American economic activity. The difference with the Chinese cohort is not headcount. It is that they treat AI as the production engine of the business, not as an occasional assistant.
Picture the same playbook in a European setting. A copywriter in Lyon who drafts, translates and localises campaigns for three markets in an afternoon. A wedding photographer in Antwerp whose culling, editing and album design run overnight while she sleeps. A bookkeeper in Utrecht who lets an AI agent pre-sort receipts and draft the quarterly VAT return for review. None of this requires venture capital or a Suzhou-style subsidy; the monthly tool budget is comparable to a coworking coffee allowance.
The playbook, translated
Start from one sellable deliverable, not from "an AI business". Wang sells finished commercials. Gao sells design work. In each case the client buys an outcome they already understand, and AI stays invisible in the engine room. The weakest solo businesses out there sell "AI services" in the abstract.
Keep the stack lean and cancellable. The Chinese founders profiled spend under €120 a month on tools and swap them freely as better ones appear. That discipline — monthly subscriptions, no long contracts, no custom builds until a workflow has proven itself — is worth copying exactly.
Mind the rules Chinese founders can ignore. A European solopreneur feeding client data into AI tools is a data controller under GDPR: check where the tool processes data and whether a proper data processing agreement is in place before uploading anything identifiable. And under the EU AI Act's transparency rules, AI-generated content such as synthetic video or cloned voices must be disclosed as such. Neither point is a blocker; both are cheaper to handle on day one than after a client asks.
Borrow the community trick. SoloNest's meetups — the group recently held its 134th — are as much a part of the model as the software. Working alone does not have to mean deciding alone, and most European cities have freelance meetups that fill the same role.
When this isn't the right move
The survivorship numbers deserve as much attention as the success stories. SoloNest's own data, cited in the coverage, shows that only about 20 percent of its members have sustainable revenue; 40 percent are still searching for steady income and another 40 percent have not launched at all. The seven million registrations say a lot about how easy starting has become, and much less about how many of these businesses will still exist in 2027.
Some situations argue for staying put. If your income needs to be predictable next month, a solo venture built on project work is a stressful vehicle, and unlike Chinese founders you will not find a €2,400 municipal grant waiting in most EU cities. If your service is exactly what AI is commoditising — generic copy, stock-style imagery, routine translation — cheap tools cut both ways, because your prospective clients have them too. And regulated work such as accounting, law or medicine still carries professional requirements that no subscription replaces.
The interesting question for most people reading this is not whether to quit and register a one-person company next week. It is narrower: which parts of the work you already do could run the way Wang's and Gao's businesses run, and which genuinely need you. That inventory is exactly what Cresly's AI Readiness Scan is built to produce — a concrete map of where AI fits in your business, and where it does not, before you spend a euro on new tools.