Kleiner Perkins has secured $3.5 billion across two new funds, a clear signal that one of Silicon Valley’s most storied venture firms is concentrating its firepower on artificial intelligence. The raise includes $1 billion for its 22nd early-stage fund and $2.5 billion for a growth vehicle, marking a sharp step up from the firm’s last fundraise and underscoring how fast AI has become the center of gravity in venture capital.
The move cements a strategy already visible in the portfolio: early stakes in Together AI, Harvey, and OpenEvidence, and positions in Anthropic and SpaceX—two companies widely watched for potential public market debuts. For a firm that made its name backing Amazon and Google, the message is familiar but timely: when a platform shift is this large, you scale into it.
A Two-Fund Engine for AI Across Early and Growth Stages
The $1 billion early-stage fund is designed to capture new technical breakouts at seed and Series A, while the $2.5 billion growth fund gives Kleiner Perkins the ballast to double down on winners as they scale. Expect an emphasis on the AI stack: frontier and foundation models, inference and training infrastructure, data and orchestration layers, and vertical applications where domain expertise matters—legal, healthcare, finance, defense, and industrials.
Investments like Together AI point to a thesis around lowering compute and deployment friction for enterprises embracing open-source and fine-tuned models. Harvey’s traction with major law firms and OpenEvidence’s push into clinical decision support illustrate where regulated, high-stakes workflows are ripest for AI-native software—if vendors can demonstrate accuracy, auditability, and clear liability frameworks.
Industry data backs the focus. PitchBook and NVCA have noted that while overall U.S. venture activity cooled and exits lagged, AI deal value and mega-rounds have remained disproportionately resilient. IDC and Gartner continue to forecast rapid compound growth in enterprise AI spending as pilots move into production, a shift accelerated by falling inference costs and improved tooling.
Returns Fuel the Next Push in AI Investment Strategy
Fresh capital often follows realized gains. Kleiner Perkins benefited from last year’s Figma IPO, where it led the company’s $25 million Series B back in 2018. The firm also saw a return from Google’s acqui-hire of portfolio company Windsurf. With public listings still scarce, distributions like these matter—they boost limited-partner confidence and free up risk appetite for new bets.
The growth fund also positions the firm to be a long-term partner to breakout AI companies that increasingly require multi-billion-dollar compute budgets and complex go-to-market motions. Having both early and late-stage capital under one roof is a competitive advantage when founders want continuity through rapid scaling and capital-intensive model development.
Lean Team Betting on Concentration in AI Dealmaking
Kleiner Perkins now operates with a tight partner bench—five investors making firmwide decisions. Recent changes include Ev Randle’s departure to Benchmark and Annie Case shifting to an advisory role. A smaller partnership can be a feature, not a bug, in AI: cycles move fast, technical diligence is deep, and winning allocations often hinge on conviction and speed rather than committee sprawl.
Concentration also reflects lessons from previous platform shifts. The firm’s history with category-defining companies suggests a preference for fewer, bigger positions in markets where power laws are extreme. In AI, where model quality, proprietary data, and distribution converge into moats, the difference between first and second place can be decisive.
What Founders Should Expect from Kleiner Perkins’ AI Focus
With this capital, founders can expect larger seed and Series A rounds for technical teams tackling hard problems, and a path to substantial follow-on for those that hit product-market fit. But the bar is rising. Investors are scrutinizing gross margins after compute costs, data provenance, model evaluation rigor, and security posture. Metrics like user retention, time-to-value, and cost per inference are moving to the front page of partner meetings.
Procurement remains a gating factor in regulated industries, making referenceability and compliance tooling—model cards, indemnities, red-teaming—part of the sales kit. Partnerships with cloud providers and chipmakers can help reduce unit costs and unlock co-selling. Companies that navigate these realities can scale quickly; those that can’t will feel the gravity of longer enterprise cycles.
The Competitive AI Map Among Multi-Stage Venture Firms
Kleiner Perkins’ posture lands amid an arms race among multi-stage firms vying for the AI crown. Dry powder across U.S. venture funds remains well above $300 billion, according to PitchBook, and a meaningful share is earmarked for AI. The contest spans closed and open-source approaches, from frontier labs to inference platforms and domain-specific copilots.
The firm’s bet is clear: AI is not a single category but the substrate of the next generation of software and infrastructure. With $3.5 billion to deploy and a track record of betting early on paradigm shifts, Kleiner Perkins is positioning itself to write the next chapter—one where compute, data, and distribution determine the winners faster than most cycles we’ve seen.
