Elena Volotovskaya is the Head of Softline Venture Partners.
The AI boom led American investors to put 75.6% more money into startups in the first half of 2025 than in the same period last year. More than half of all startup funding went to AI projects.
In monetary terms, over $104 billion out of $162.8 billion flowed into AI. Venture funds clearly see huge potential in the new wave of tech companies. But much of this investment is driven by expectations rather than actual revenue. That creates the risk of sharp corrections that could make your portfolio volatile.
As the head of a corporate venture fund, I view AI as both a generational opportunity and a sector where due diligence must go far beyond surface numbers. Here are five factors shaping valuation multiples in AI startups and how to evaluate a company you’re considering financing.
1. Multiples
Classic P/S, P/E and EV/revenue don’t work for valuing AI startups. That’s because many AI projects are still in the early stages and don’t have stable revenue or profits. Investors pay for potential, technology and unique datasets.
Relying on average market benchmarks is also tricky: High-profile startups skew the picture with extremely high multiples. For example, Perplexity’s valuation-to-revenue multiple is around 100x—the company raised $200 million at a $20 billion valuation, according to TechCrunch. Most vertical AI startups operate at much lower levels—between 10x and 40x, according to FINRO.
When evaluating AI startups, look beyond headline multiples and consider the context, business model and growth potential of each company individually.
2. Compute Power
When valuing AI startups, it’s also important to consider the costs of compute resources. Training LLMs requires enormous amounts of energy, servers and GPUs. A McKinsey study shows that by 2030 “data centers are projected to require $6.7 trillion worldwide to keep pace with the demand for compute power. Data centers equipped to handle AI processing loads are projected to require $5.2 trillion in capital expenditures.”
In any case, demand for compute power today exceeds supply. Access to GPUs and energy is expensive, and startups are dependent on major players that control that access like Microsoft, Google and Amazon.
When evaluating and analyzing a startup, take into account cloud contracts, compute partnerships and the sustainability of the company’s infrastructure strategy.
3. Intellectual Property
Unique datasets and proprietary model architectures increase a startup’s value and can justify high multiples. But in the case of AI startups, data—or rather the rights to it—can become a potential legal issue. Companies are already accusing AI giants of copyright violations.
For example, a lawsuit against OpenAI and Microsoft, initiated by the New York Times in 2023, has been ongoing. The media outlet claims that AI developers used their content without permission to train models and generate responses.
At the same time, major AI players are striking deals with media brands. For example, Microsoft is looking to build an AI marketplace through which publishers can earn revenue from the use of their content.
Either way, the use of data by AI agents is poorly regulated. For investors, it’s crucial to assess whether a startup can reliably protect its intellectual property, or if legal uncertainty could suddenly erode its multiple.
4. Regulation
Legal regulations are being developed not only around data but also around AI development, its usage and interaction with users in general. The first EU artificial intelligence law came into effect in 2024, and in early 2025, the European Commission introduced an updated version. The AI Act classifies AI products by risk level and imposes different requirements depending on the model and the area of use. Products used in healthcare, education and other sensitive sectors face higher standards for quality and transparency.
At the same time, the U.S. “relies on existing federal laws and guidelines to regulate AI but aims to introduce AI legislation and a federal regulation authority,” according to White & Case.
In other words, legislation is fluid and evolving, and this factor also affects a startup’s performance and, consequently, its valuation multiple.
5. The Global Funding Gap
Despite the global boom around artificial intelligence, the U.S. remains the dominant investor in AI projects. The U.S. private sector invested more than $100 billion in AI in 2024—about 10 times as much as second-ranked China.
European countries are far behind. The U.K., which ranks third in AI investment, put roughly $3.9 billion into British AI companies last year.
Strict regulations and a more conservative investment culture have pushed European company valuations lower compared to their U.S. peers. As a result, multiples in Europe and other regions are significantly lower.
This creates both opportunities and risks. Strong teams in Europe or Asia may be available at more reasonable valuations. But at the same time, global AI leadership risks becoming concentrated in just a few markets, leaving others to play catch-up unless their local capital ecosystems mature.
Balancing Risk And Reward
AI startups can look like risky bets. The huge flow of money into AI projects makes the space feel like a bubble, often compared to the dot-com era. A recent MIT study even found that 95% of AI pilot projects fail to deliver meaningful results.
That said, the AI bubble isn’t empty. Some of the biggest players are already generating massive profits (paywall): “Microsoft’s Azure cloud service, heavily focused on AI, grew 39% year-over-year to an $86 billion run rate. OpenAI projects $20 billion in annualized revenue by the end of the year, according to The Information, up from around $6 billion at the start of the year,” as Fortune reports.
Investing in AI comes with risks, just like any other investment. But with a comprehensive analysis and careful evaluation of an AI startup, investors can realistically assess the risks and opportunities and make a data-driven decision.
Forbes Finance Council is an invitation-only organization for executives in successful accounting, financial planning and wealth management firms. Do I qualify?
