The a16z Show: "The State of Markets"

Date: February 9, 2026
Host: Andreessen Horowitz (a16z team)
Featured Guest: David George, General Partner

Overview

This episode of The a16z Show delivers a data-rich, inside look at the explosive growth, turbulence, and future trajectory of AI-driven markets and companies. General Partner David George deep-dives into revenue trends, efficiency metrics, market impacts, enterprise adoption struggles, company case studies, and macro investment themes. Listeners gain both a high-level overview of AI’s transformation of software business and detailed anecdotes from the a16z portfolio.

Key Discussion Points & Insights

1. Historic AI Growth & Demand

• AI Outpaces SaaS: The fastest AI companies are reaching $100M in revenue at unprecedented speeds, growing an astonishing 693% year over year in 2025. They’re hitting high revenues with notably less sales and marketing spend than previous SaaS standouts.
  • “The best AI companies that are growing the fastest are not the ones spending the most... They’re spending less... yet growing much, much faster.” (B, 04:21)
• Product Demand, Not Efficiency: Unlike past booms driven by efficiency plays or spend, the AI surge is above all demand-driven. These companies “can barely keep up on the supply side; every GPU that gets plugged in is maxed out immediately.” (A, 00:21)

2. Company Margins & Operational Changes

• Gross Margins as an AI Badge: AI companies have lower gross margins—seen as a sign of genuine product use, not just better cost structure.
  • “Low gross margins for AI companies are sort of a badge of honor, in the sense that we want to see if low gross margins are a result of high inference costs. That means people are using AI features.” (B, 05:22)
• New Efficiency Metrics: AI firms are achieving astonishing ARR (Annual Recurring Revenue) per employee, up to $1M/FTE, compared to SaaS’s ~$400k/FTE.
  • “For the best AI companies, they’re running at $500,000 to a million dollars per FTE... mostly because demand is very, very strong for their products.” (B, 05:59)

3. Adapting or Getting Left Behind

• AI-Native vs. Legacy Firms: Companies must “adapt to the AI era or die.” It’s about embedding AI at the product core, not “just attaching a chatbot.”
  • “You need to adapt to the AI era or die... Be aggressive about disrupting yourself and changing.” (B, 07:13)
• Rapid Change in Engineering: Switch to modern AI coding tools is yielding 10-20x productivity gains. That means entire org structures in engineering, product, and design will be fundamentally rethought within a year.
  • “He thinks it’s going somewhere between 10 and 20x faster than progress before.” (B, 08:36)
  • “For every task... can I do it with electricity or do I need to do it with blood? This is like the extreme mindset shift.” (B, 09:56)

4. Transformation of Business Models

• Spectrum of Evolution: From licenses → SaaS/subscription (seat-based) → usage/consumption-based → outcome-based.
  • “The next iteration will be outcome-based... The only area where that’s really possible today is probably customer support, customer success... If other functions can measure those kinds of outcomes, that would be a huge disruptive force.” (B, 11:46)

5. Company Case Studies & Engagement Data

AI in Action

• Harvey (Legal AI): Lawyers are using the product roughly double previous engagement levels.
  • “Turns out lawyering and reasoning go hand in hand... Users are spending about double the amount in the product as they had before.” (B, 15:37)
• Abridge (Healthcare AI): Doctors rave about the time and life improvements. As user count multiplies, engagement stays high or improves.
  • “They have extremely high usage among the people who use the product, and that has actually held steady and grown a little bit even as they've added tons of more users.” (B, 17:07)
• 11Labs (Voice AI): Voice as a foundational input. Explosive growth and high efficiency.
  • “The usage growth... is just staggering. This company is growing very fast and it’s a great example.” (B, 18:17)
• Navan (Travel AI): AI now handles 50% of customer interactions, resulting in 20-point gross margin improvement over three years.
  • “Their competitors are not adapting... while they’ve been doing things the old way, Navan now has 20 percentage point higher gross margins.” (B, 19:24)
• Flock (Crime prevention AI): Solving 700,000+ crimes a year, with nearly 10% more crimes cleared per officer where deployed.
  • “What their ROI is, is solving crime. Each year, Flock is solving 700,000 crimes.” (B, 20:36)

6. Enterprise & Fortune 500 Adoption

• Change Management Bottleneck: Fortune 500 CEOs say they want AI, but actual org change is slow and hard.
  • “What we're hearing from Fortune 500 CEOs is... ‘We're ready to change, we’re going to become AI companies.’ That’s quite different than what is actually happening... Change management is hard.” (B, 21:47)
• Huge productivity wins for some: Chime cut support costs by 60%. Rocket Mortgage saved 1.1 million hours in underwriting—$40M/year.
  • “There’s going to be a reckoning over the next five years of who can actually embrace change.” (B, 23:23)

7. Public & Private Market Trends

• Power Law Concentration: Both public and private markets see value concentrating even more into outlier companies. 10 largest unicorns constitute 40% of total unicorn value—double since 2020.
  • “Value very much concentrates in the outlier companies... the collective valuation... about five and a half trillion dollars. The ten largest ones comprise almost 40%...” (B, 41:03)
• Market Dynamics: High growth and high margin companies receive major valuation premiums. Unlike previous bubbles, current CapEx is driven by highly profitable giants.
  • “The underlying fundamentals bear little resemblance to previous bubbles. The investment is financed primarily by... profitable companies.” (B, 25:09)

8. Supply Side: CapEx, Debt, and Buildout

• CapEx Built on Financial Strength: AI infrastructure buildout is massive yet so far supported by industry profits, not just speculation.
  • “Capex is actually supported by cash flows and capex as a percentage of revenue is considerably lower [than dotcom].” (B, 28:44)
  • “There is no dark GPU... you put a GPU in the system, it gets fully utilized immediately.” (B quoting Gavin Baker, 33:40)
• Risks Emerge: Debt now entering as cashflow can’t keep up with AI infrastructure needs (e.g., Oracle making outsized bets).

9. AI Revenue and Market Impact

• Where is AI revenue now?: Current annualized AI-enabled revenue estimated at ~$50B, but expected to hit $1T by 2030 (~1% global GDP).

  • “We’re probably in the... $50 billion range... but it’s growing way, way, way faster than 100% year over year.” (B, 39:17)

• Public Market Impact: AI winners are driving ~80% of the S&P 500’s return; price increases in the market have fundamental support rather than pure froth.

Notable Quotes & Memorable Moments

• On adapting to AI:

  “You need to adapt to the AI era or die... Your employees are now your AI agents. How many agents do you have?”
  (B, 09:56)

• Extreme rethinking of operations:

  “For every task... can I do it with electricity or do I need to do it with blood?”
  (B, 09:56)

• Efficiency mindshift:

  “The big debate... was like, oh, my gosh, look at the AI efficiency gains... To wholesale run your company totally differently, I think... we’re kind of early in that journey.”
  (B, 12:49)

• AI in law:

  “LLMs have actually increased my workload because every client thinks they’re a lawyer now.”
  (C quoting tweet, 14:57)

• AI supply chain:

  “There is no dark GPU... you put a GPU in the system... it gets fully utilized immediately.”
  (B quoting Gavin Baker, 33:40)

Important Timestamps

• AI’s revenue and growth outpacing SaaS (00:00 – 05:59)
• Measuring efficiency – ARR per FTE (05:59 – 07:13)
• AI impact on pre-existing companies – adapt or die (07:13 – 10:54)
• Business model evolution: licenses → SaaS → usage → outcome-based (10:54 – 12:44)
• Company spotlights: Harvey, Abridge, 11Labs, Navan, Flock (15:08 – 21:05)
• Enterprise adoption & change management (21:12 – 24:44)
• AI power laws and market trends, public/private (40:43 – 44:55)
• Databricks and embedding AI (44:55 – 46:49)

Final Takeaways

• We’re still (very) early: The product cycle for AI is just beginning; the fundamentals driving the market look robust and risk is mainly in companies failing to adapt.
• Demand, not spend or efficiency tricks, is driving AI’s rise.
• Massive company and workflow overhaul is ahead: org structures, business models, and even definitions of employee “agents” will shift.
• Concentration accelerates: Winners take most in value accrual, making adaptation even more critical for companies of all sizes.
• The next 12-24 months are pivotal: Expect more dramatic transformation and clear separation between the AI-haves and have-nots.

Panelists maintain an analytical yet candid tone, sharing optimism about innovation and disruption while openly discussing risks and uncertainties.

(This summary omits podcast intro/outro advertisements and solely covers the in-depth content.)