The New Rule for Picking AI Winners | The a16z Show

2026-05-29 · Source: a16z · Field: Finance & Economics — Capital Markets & Investment Management, Economic Analysis & Policy, Entrepreneurship & Start-ups · Depth: Intermediate, extended

Summary

The AI market is experiencing unprecedented growth, with Anthropic and OpenAI projected to reach a \$200 billion revenue run rate by year-end, surpassing monthly revenue additions of Meta, Google, or Microsoft. Top 1% exit valuations have surged from \$10 billion in 2020 to \$32 billion recently, potentially exceeding \$100 billion by September. Despite this rapid expansion, AI technology diffusion into the broader economy remains under 5%, indicating vast untapped potential beyond current tech-forward applications like coding. Enterprises are expected to allocate significant portions of their \$2 trillion annual profits to AI, driving demand. The industry is shifting towards native, proactive AI applications, but the rapid pace of change challenges defensibility, with 40% of Forbes AI 50 startups dropping off in a year. Value capture hinges on the "token path" and the competitive market structure of foundational model providers, with cheaper, slightly less capable models (e.g., Chinese LLMs, 10x cheaper) posing a dilemma. Current AI valuations are not considered a bubble due to severe supply constraints in compute, data centers, and power, which are expected to persist for 3 years. This influx of hypergrowth AI companies is anticipated to invigorate public markets, offering new high-growth investment opportunities beyond the "Mag Seven" (growing sub 30%).

Key takeaway

For investors evaluating AI opportunities, recognize that the market's hypergrowth and severe supply constraints (e.g., data center capacity until late '28) create a unique environment, not a typical bubble. Your strategy should prioritize companies deeply integrated into the "token path" and anticipate rapid shifts in market leadership, as evidenced by 40% of top AI startups dropping off annually. Be prepared for higher loss ratios in early-stage investments, focusing on backing leading founders in promising, high-risk areas.

Key insights

The AI market is undergoing hypergrowth and massive value creation, driven by foundational models despite early diffusion and supply constraints.

Principles

• Value accrues significantly to companies in the "token path."
• Market structure of frontier model providers dictates token pricing.
• Rapid technological shifts erode defensibility for early movers.

In practice

• Monitor "token path" companies for value capture.
• Evaluate AI investments considering supply constraints.
• Anticipate rapid shifts in market leadership.

Topics

• AI Market Dynamics
• Venture Capital Strategy
• AI Valuations
• Supply Chain Constraints
• Generative AI Applications
• Value Capture
• Open-Source AI

Best for: Investor, Entrepreneur, Consultant

Related on AIssential

• AI Markets: Deep Dive with a16z's David George — AI-native companies exhibit unprecedented growth and efficiency, driven by strong demand and early product cycle dynamics.
• AI startups now make up nearly half of US venture capital market: Report - Business Today — AI startups dominate US venture capital, driven by mega-rounds and creating a significant valuation gap.
• Google Cracked AI Monetization. Meta, Microsoft, and Amazon Haven't. — Clear AI monetization strategies that demonstrate value creation beyond efficiency are crucial for investor confidence.
• Are we tokenmaxxing our way to nowhere? — The AI sector is experiencing rapid, exclusive growth, creating a significant divide between insiders and the public.
• The bubble is slowly popping, investment isn't able to keep up — The AI market faces a critical debate regarding its investment sustainability and the long-term viability of current pricing models.
• Sam Altman and the day Nvidia’s meteoric rise came to an end — The "scaling = AGI" myth's collapse led to a tech stock plateau, revealing LLMs' commodity nature.

Open in AIssential →

Editorial summary, takeaway, and curation by AIssential. Original article published by a16z.