A seismic shift is on the horizon for the tech industry, with venture capital luminaries Jason Lemkin, Harry Stebbings, and Rory O’Driscoll forecasting a monumental year for initial public offerings (IPOs) in 2026. During a recent 20VC x SaaStr discussion, the experts delved into major tech deals, evolving valuations, and the disruptive potential of AI, predicting a staggering $1.4 trillion IPO wave that could reshape the investment landscape.
The discussion covered a wide range of topics, from SpaceX's formidable valuation and Netflix's market dominance to the comeback of Tiger Global and the looming platform risk for AI applications. Here are the key takeaways for founders and investors:
Top Insights from the 20VC x SaaStr Discussion
1. SpaceX's $800 Billion Valuation and the New Normal of Down Rounds
SpaceX is reportedly pursuing an $800 billion valuation through secondary sales, despite generating roughly $15 billion in revenue with approximately 30% growth. This implies a valuation north of 40 times its run rate. While this might seem aggressive, the experts highlighted a significant shift in market sentiment: "Every single IPO in 2025 was a down round from the prior private round. And nobody cared," noted Jason Lemkin. Rory O'Driscoll added, "Every time that private market valuations came into contact with public market valuations, private market valuations were found wanting." The angst surrounding overpriced private rounds has dissipated, with founders now less concerned about raising at lower valuations in subsequent rounds.
2. 2026: The Mother of All IPO Years
If tech giants like SpaceX, Anthropic, and Databricks all go public in 2026, the market could see an unprecedented surge. This scenario could generate approximately $1.4 trillion in market capitalization and potentially return $700 billion to venture capitalists. While this represents only about 20% of the total private venture fair market value (estimated at $2.9 trillion), it would still mark the best year ever for venture liquidity. The math is simple: a single SpaceX IPO at $400-600 billion would define the entire year, much like Facebook did in 2012 and Alibaba in 2018.
3. Anthropic's Predicted $350-500 Billion IPO
Anthropic, an AI leader, has maintained a "remarkably sober-minded and sensible" financial plan, focusing on efficiency and lower burn rates. If the company achieves $25-26 billion in revenue next year (a 3x growth), a 20x multiple would place its valuation at $500 billion. This multiple aligns with Databricks' current trading valuation and is not considered aggressive for top-decile growth companies. Jason Lemkin's bet: a $500 billion Anthropic IPO in the latter half of 2026.
4. Netflix's Dominance and the Next Industry to Be "Eaten"
Netflix's market cap of $470 billion dwarfs even the largest traditional studios, which are typically under $200 billion. This dominance suggests that Netflix could acquire a major player like Warner Bros with less than 20% dilution, continuing its expansion. This trend exemplifies how venture-backed, tech-enabled companies disrupt old-school industries, much like Google and Facebook transformed advertising, and Amazon revolutionized retail. The critical question for founders now is: "What's next?" Fintech is a strong contender, with companies like Revolut potentially growing to become Europe's largest market cap banks and acquiring legacy players.
5. Harvey's $8 Billion Valuation: A Bet on Growth, Not Operations
AI legal assistant Harvey recently raised $160 million at an $8 billion valuation, representing less than 2% dilution. With $150 million in Annual Recurring Revenue (ARR), 300% year-over-year growth, 98% Gross Dollar Retention (GDR), and 168% Net Dollar Retention (NDR), the metrics are compelling. Jason Lemkin explained the simple math: if Harvey triples its ARR to $450 million next year, investors are effectively paying 20x forward revenue. Rory O'Driscoll highlighted the core risk: "When you're paying $8 billion, there's only one risk you're running. You have no operational risk. You have no business model risk. You have no team risk. The only risk you're running is you're paying eight for something that might be worth four."
6. Tiger Global's $2.2 Billion Fund and the "Truth Serum" of Money
Tiger Global's new $2.2 billion fund, featuring a 20% General Partner (GP) commitment (approximately $400 million of their own money), signals a deep commitment. This level of personal investment, pushing the limits of pure economic viability, acts as a "great truth serum," as Rory O'Driscoll put it: "Don't tell me what you think, tell me what you do." This move by Chase Coleman is seen as an effort to rebuild trust after the challenges of 2021.
7. The "Jaspered" Risk: AI Apps' Platform Instability
A significant concern is the potential for current AI applications to become obsolete due to rapid advancements in deep reasoning. Today's AI apps rely on relatively simple prompts, delivering quick answers. However, as deep reasoning capabilities evolve from minutes to seconds or milliseconds, the landscape could drastically change. "The Harvey we use in 2025 could seem quaint in two years," warned Jason Lemkin. Just as ChatGPT disrupted early AI leaders like Jasper, a new generation of AI models could emerge, making today's "locked-in" applications vulnerable to disruption.
8. LLMs as Commodities: Value Through Consolidation
Mark Benioff's assertion that Large Language Models (LLMs) are commodities holds some truth; swapping between Anthropic, OpenAI, and Gemini is easier than switching cloud providers like AWS, Azure, and GCP. However, commodity markets can still foster successful businesses through consolidation. Examples include the hard drive industry, where a few survivors thrive, and airlines, where consolidation creates value despite high fixed costs and easy switching. The key question remains whether LLM providers will consolidate into a few profitable players or if open-source models, particularly from China, will keep the market fragmented.
9. Chinese Open-Source Models Gaining Traction
Martin Casado of Andreessen Horowitz (a16z) revealed that approximately 20% of a16z portfolio companies utilize open-source models, and of those, 80% opt for Chinese open-source solutions like DeepSeek. This highlights a significant latent demand for affordable and capable open-source models, especially since Meta has scaled back its state-of-the-art open-source contributions. Chinese tech companies are strategically filling this void. The prevailing sentiment is that if there are no security issues and it's legal, US startups will use the cheapest viable option, provided data is not sent back to China.
10. The Stunning Airwallex vs. Ramp Valuation Gap
A striking disparity exists between the valuations of Airwallex and Ramp, two businesses with roughly similar operations. Airwallex, with $1 billion in ARR, holds an $8 billion valuation (8x multiple), while Ramp, also with approximately $1 billion in ARR, commands a $32 billion valuation (32x multiple). This 4x difference is dramatic, even considering potential concerns about China exposure for Airwallex. The gap is further emphasized when comparing it to Brex, valued at $13-14 billion with worse growth and $400 million less ARR.
A Potentially Staggering 2026
The venture capital world is poised for what could be its greatest liquidity event in history. The anticipated IPOs of SpaceX, Anthropic, and Databricks in 2026 promise to return an unprecedented amount of capital to Limited Partners (LPs). However, this period of immense opportunity also brings significant shifts. The rapid evolution of AI could disrupt established applications, LLMs are becoming commoditized, and Chinese open-source models are gaining market share, introducing new geopolitical considerations. As the experts concluded,
