Venture Capital
Venture capital is money that investors put into very young companies in exchange for equity stakes in them. It funds ventures that may fail but can grow enormously if successful — which is why almost the entire AI industry depends on it.
A newly founded company needs money at the start but doesn’t yet earn any. It usually can’t get a bank loan either, because it owns nothing it could put up as collateral. Venture capital is the answer to this problem: investors give the company money and receive a stake in it in return. So they don’t get fixed interest back, but instead own a piece of the company. If the company goes bankrupt, the money is lost. If it grows big, the stake eventually becomes worth a multiple of its original value. The English term for this is venture capital, or VC for short.
Why loss is factored in here
Venture capitalists expect that most of their investments will bring in nothing. A typical rule of thumb in the industry: out of ten funded companies, about seven fail, two do okay, and one becomes very large. That one has to save the entire fund. This sounds like a bad deal, but it works because a loss ends at a maximum of 100 percent, while a gain can be 50-fold or 100-fold.
This calculation leads to a way of thinking that often seems strange from the outside. An investor doesn’t look for the company with the best chances of success, but for the one with the biggest possible outcome. A solid company that reliably grows ten percent a year is uninteresting to them. That’s why venture capitalists push their companies toward aggressive growth, often at the expense of profit.
This is crucial for the tech industry. Almost all the well-known names of today — from Google to Airbnb to OpenAI — started with venture capital. Without this form of financing, many of these companies wouldn’t exist, because their first years were pure loss-making years.
Rounds, valuation, and dilution
The money doesn’t come in one lump sum, but in rounds. The first is often called the seed round, involving smaller sums for an initial version of the product. This is followed by Series A, Series B, Series C, and so on, each with larger amounts. Before each round, what the company should be worth overall is negotiated. This figure is called the valuation and is not a measured value but a price the two sides have agreed on.
An example: a company is valued at 20 million euros and raises 5 million. Afterward, it is calculated to be worth 25 million, and the new investor holds one-fifth. The founders' shares shrink proportionally in the process — this is called dilution. It’s not automatically bad: a smaller stake in a much bigger company can be worth significantly more.
At the end comes the exit. Either the company is bought by a larger corporation, or it goes public on the stock exchange. Only then does the stake turn into real money. Until that point, the profit exists only on paper, and funds often wait eight to twelve years for it.
Venture capital in AI headlines
When a news report says an AI startup has “raised 300 million at a valuation of 4 billion,” it’s about venture capital. It’s important to understand: the 4 billion doesn’t exist anywhere. It’s an extrapolation based on the price of the last round. If market sentiment drops, the next round can take place at a lower valuation. This is called a down round and is considered a bad sign.
In the AI industry, the sums are unusually high because training large models requires data centers full of expensive chips. Such costs arise before a single customer pays anything. That’s why amounts that used to be enough for entire funds now flow into a single company.
A common misconception is confusing venture capital with private equity. Private equity funds buy established companies with ongoing revenue, often entirely and financed with debt. Venture capital buys small stakes in companies that can barely show anything yet. Both are forms of equity capital, but the risks and time horizons are completely different.