Menlo Ventures closed $3 billion across two new funds, its largest raise in the firm’s 50-year history, with AI as the sole investment thesis. The capital is split between Menlo Ventures XVII for seed and Series A and Menlo Inflection IV for Series B and beyond, according to Business Insider.
The Anthropic Bet That Reshaped the Firm
The raise arrives on the back of one decision. Menlo first invested in Anthropic in 2023, when the company was pre-product and pre-revenue. Most investors at the time viewed the large language model race as effectively over. Menlo saw a team building a platform, not a chatbot, at a moment when enterprises, banks, and governments needed an AI vendor that was not tied to Microsoft or Google, TechCrunch reports.
That stake is now worth approximately $14 billion. Anthropic has filed plans for a 2026 IPO at an expected target of $1 trillion or more, according to Memeburn. If the IPO lands at that valuation, it would represent the largest exit in Menlo’s history by a wide margin.
Capital Is No Longer the Constraint
The broader signal is structural. Companies spent $37 billion on generative AI in 2025, a 3.2x year-over-year increase, according to Menlo’s own research cited by Memeburn. Venture firms, crossover funds, corporate strategic investors, and sovereign wealth funds are all competing for AI deals. When capital becomes this common, it stops functioning as a competitive advantage.
Menlo partner Venky Ganesan framed the shift directly: being close to the companies defining the category is an advantage that capital alone cannot buy, Memeburn reports. Investors who built information networks and pattern recognition during the early AI years hold a structural edge that latecomers cannot easily replicate.
What Differentiates Winners Now
The convergence problem compounds the difficulty. Nearly every major AI model has some real technical edge: speed, cost, coding benchmarks, reasoning, distribution, or ecosystem integration. As capabilities converge, the differences that separated leaders from the pack are narrowing. A product that looks dominant today can lose tomorrow to a cheaper alternative bundled into a larger platform, according to Memeburn.
The question investors are now asking is whether a company owns something that cannot be copied. Proprietary data and platforms that others build on top of keep surfacing as the durable answers. Everything else, including model performance on benchmarks, is temporary.
For teams building in the agent stack, the capital environment is favorable but the selection pressure is intensifying. Money will find AI companies. Whether those companies can hold their position once cheaper models and bigger platforms arrive is the bet Menlo is making with $3 billion.