The AI venture capital ecosystem is poised for dramatic upheaval in 2025 due to unprecedented investment scale and technological innovation. AI startups raised approximately $100 billion (about 144 trillion won) in the first half of the year alone, accounting for nearly half of all global startup investment. Funding is concentrated in massive deals across various sectors, including AI technology, infrastructure, and applications, and related companies are rapidly emerging.
In light of this trend, Crunchbase met with six leading AI venture capitalists and compiled their insights on market structure, startup investment strategies, and the next generation of growth drivers.
Accel partner Philippe Botteri points out that the six largest U.S. companies—Nvidia (NVDA), Microsoft (MSFT), Apple (AAPL), Alphabet (GOOGL), Amazon (AMZN), and Meta (META)—are pouring hundreds of billions of dollars in cash into AI infrastructure. However, he emphasizes that despite this massive capital investment, startups based on core AI technologies still have opportunities to explore new areas. In fact, Accel has proactively invested in AI companies covering both the model and application layers, such as Anthropic, Perplexity, Synesthesia, and Sierra.
Steve Barthallo, a partner at Foundation Capital, analyzed that the bottleneck in the AI industry is gradually shifting to physical infrastructure, and the hardware ecosystem based on chips, power, data centers, etc. will become the core opportunity in the future. Leveraging its experience in incubating AI chip manufacturer Cerebras early on, Foundation Capital has invested in more than 100 AI startups, recently investing in Tenor, a company that automates medical licensing, and PlayerZero, a tool that provides coding error prediction.
Dell Technologies' venture capital arm, DTC, while selling GPU servers, is also actively investing in AI chips and application software. Daniel Doctor, Managing Director of DTC, stated, "Investment in AI startups is happening faster than ever before," and revealed that some deals were finalized within two days of a meeting. Dell has also begun the process of acquiring AI chip startup Rivos, which is currently awaiting regulatory review.
Tim Gullari, CEO of Sierra Ventures, points out, "If computing power is the bottleneck, then data is the differentiator." Sierra divides its AI investments into five tiers, from infrastructure to vertical applications, with a particular focus on the application tier where unique data and distribution strategies are crucial. Gullari states that of the world's $110 trillion GDP, the majority is still concentrated in the service sector, which has not yet been digitized by AI, and the value creation based on AI is expected to explode in the coming decades.
Andrew Ng, co-founder of the AI Foundation and Google Brain, has chosen a strategy of exclusively acquiring data through a venture studio model. His fund partners with large companies such as AES, HP, and Mitsui, designing AI companies based on data from real-world industries and directly hiring CEOs to launch these ventures. Ng emphasizes, "AI is not a single technology, but a network that creates customized opportunities across various industries."
Finally, Alphabet's investment arm, GV, boldly invests even in AI startups that may conflict with Google's existing businesses. GV provides funding from early stages, from chips and compilers to user applications, and accepts high-premium valuations based on rapid revenue growth and market opportunities.
This shows that the key words that major venture capital firms, from Accel to Sierra Ventures and GV, are focusing on are "computing resources," "unique data," and "application innovation." The AI boom is not just a simple trend, but a technological and industrial revolution that is very likely to become a decisive variable in giving rise to new unicorn companies in various industries in the coming years.
