Are venture capital fund managers making themselves obsolete in their search for the next AI unicorn?

Leading players in the market have long recognized this: “We are in the last ten years of venture capital as we know it,” says James Currier, founder of the successful seed-stage investment firm NFX. The irony is that it is the very technology financed by VCs – artificial intelligence (AI) – that is now disrupting the industry. “The venture capital sector, in particular, is on the brink of a significant transformation,” says Tina Dreman, 2023 German Investor of the Year.

Of course, the technical innovations taking place in artificial intelligence are so phenomenal that making reliable predictions has become increasingly difficult. However, we are clearly in the midst of a true technical revolution. The past year brought us smart AI chatbots which are already an integral part of many day-to-day processes. It gave us AI artists whose works are worth millions: AI agents that schedule our appointments, and the AlphaFold 2 AI system that helped Demis Hassabis, head of Google’s AI team, win the Nobel Prize in Chemistry.

Given the breathtaking capabilities of today’s leading AI models, the forecasts for their continued development are nothing short of exponential. Following this logic, it may take just months rather than years before we reach a point where intelligence is infinitely scalable and almost free. So how do venture capital managers fit into a world where ground-breaking, intelligent models can make investment decisions?

So far, things are going well. As early as 2023, research firm Gartner predicted that 75% of all VC managers would be using AI in their investment processes. Today, this figure seems relatively low, as AI is widely used to automate data analysis, enhance efficiency in data processing, identify patterns in start-up data, and position companies within current and projected market trends. AI tools have proven themselves to be especially useful in forecasting and scenario modeling. At the most basic level, many firms still use widely available generative models such as ChatGPT (or more recently, DeepSeek) to analyze simple datasets. Some companies, such as Earlybird, have taken things a step further by developing proprietary technologies. Earlybird’s ‘Eagle Eye’ AI is setting new standards in identifying promising startups – for example, it played a key role in the investment process for the German AI firm Aleph Alpha.

The intermediaries are also keeping pace. The PitchBook research and analysis platform recently launched its ‘VC Exit Predictor,’ which ranks the most successful start-ups based on their predicted success.

However, today’s AI models are still generally more of a tool than a complete solution. While they evidently enhance the modeling of opportunities and risks, and improve data-driven decision-making, good venture capitalists need more than just AI. Samuel Widmann, entrepreneur and founder of Endoxon, the Swiss company that was behind Google Maps and later acquired by Google, argues: “AI will not replace VCs, but it will provide rapid and crucial decision-making insights. In the end, however, it is people who make the decisions. For good investors, gut instinct always plays a significant role in the decision-making process.”

So why are industry leaders concerned? If we take the efficiency argument to its logical conclusion, AI’s growing use certainly suggests that human analysis will become increasingly redundant wherever large datasets are available. This is particularly true for late-stage transactions from Series B onward and for investors who prefer these late-stage deals.

Furthermore, when it comes to returns, AI may not be lagging as far behind human investors as you might think. An experiment published in the Harvard Business Review compared the returns of an AI-driven VC investment portfolio with those of 255 early-stage angel investors. The AI won, delivering an annual return of 7% compared to the angel investors’ 2%.

The reasoning makes sense: if quantitative hedge funds can entrust billion-dollar trades to computers, why shouldn’t algorithms work just as well in the venture capital sector? 

The human factor

Success in the venture capital industry has long been based on the same ingredients: a sharp mind, hard work, a strong network of investors and founders alike, operational and financial expertise, a pinch of luck, a keen instinct, and an unrelenting drive for success. Perhaps it is not so easy to remove the human factor from the venture capital equation. Especially in a world that has high risks on one side and limited data on the other, intuition serves as the compass for navigating unpredictable situations. The human ability to look beyond the numbers and explore dreams alongside entrepreneurs has yet to be replicated by machines. Additionally, networking, mentoring and interpersonal knowledge transfer remain fundamental to the VC role.

The human factor remains crucial in venture capital, says Philippe Bubb, founding partner of session.vc: “Founders with outstanding projects choose their investors, not the other way around. To be part of these investment rounds, you need a strong network, a convincing track record, and deep personal dedication – AI alone is not enough.” 

AI has so far shown itself to be more of a high-performance tool than a replacement. The future is likely to bring a fusion of AI and human decision-making, as this is hardly conceivable without the human factor. One thing is clear: the VC industry will undergo dramatic change in future years, and the technological revolution will serve to increase market competition. In an already highly fragmented landscape, the investors who succeed will be those who best integrate AI into their processes. In a world where the marginal cost of intelligence is approaching zero, AI has the potential to shake the very foundations of the power law – the principle that only the most successful will continue to succeed. This means that smaller venture capital firms will have greater chances of succeeding in the fierce competition to discover the next unicorn.