Imagine you are starting a company at a moment when software can draft code, organize files, summarize research, and help decide what to build next. That is the practical appeal behind the “tiny team” trend now circulating in Silicon Valley. The New York Times reported on March 28, 2026, that some founders are embracing teams as small as two: one person plus artificial intelligence. The idea matters because it aligns with a wider investment climate in which AI has become a dominant business theme. Reuters reported in January that almost two-thirds of global venture capital funding in 2025 went to AI companies, showing how strongly money and attention are flowing toward AI-shaped work.
The people most directly involved are startup founders, small product teams, and investors looking for faster ways to build companies. In the Times report, Dan Shipper of the media startup Every described a “two-slice team,” his term for one employee working with AI tools rather than a traditional cluster of colleagues. He said several products at Every are each managed by one employee, with help from coworkers when needed. Y Combinator has also signaled interest in this model from the investor side. In its Spring 2026 Requests for Startups, the accelerator said AI-native companies can now be built faster, cheaper, and with more ambition than before.
This trend is appearing most clearly in software and internet businesses, where work can be expressed as prompts, documents, code, and digital workflows. It is most useful when a team is trying to move from idea to product without adding many people or much coordination overhead. Like a pocketknife replacing a toolbox for a small repair, AI can reduce the amount of equipment needed for an early task. Public sources also place the trend in a precise time frame: the Times article was published on March 28, 2026, while Y Combinator’s Spring 2026 guidance describes the shift as something already happening, not a distant forecast.
In practice, the model works by shifting routine execution toward AI while keeping human judgment at the center. Y Combinator’s current startup requests describe tools such as Cursor and Claude Code as helpful for building software once a team knows what needs to be built. The same YC page argues that product work is changing because agents can increasingly take the first pass at implementation. Stripe has framed the commercial backdrop in similar terms. In its 2025 Sessions keynote, the company said AI is one of the two major tailwinds shaping the next decade of commerce. That does not prove every tiny team will succeed, but it does show that large platforms expect AI-assisted operating models to matter.
What happens next is likely to be a mix of opportunity and pressure. Supporters argue that a single worker can now produce more than before, and the Times article says Every used its AI-heavy structure to expand to more than 20 employees overall. At the same time, Reuters reported in March 2026 that economists and investors are growing more concerned about AI-driven disruption, with Goldman Sachs economists estimating 5,000 to 10,000 monthly net job losses last year in the most exposed U.S. industries. The clearest next step for a reader today is simple: examine one workflow that is repetitive, digital, and time-consuming, then test whether AI can handle a first draft while a human keeps final responsibility.