a16z, Anish Acharya: Is SaaS Dead? Do Margins Still Matter? Why We Are Not in an AI Bubble?
Anish Acharya, a GP at a16z, challenges the narrative that SaaS is dying and argues that the market is actually oversold on the threat of AI disruption. Rather than "vibe coding" away traditional enterprise software, he contends that AI will drive productivity gains across new use cases and that application layer companies building on foundation models will capture significant value through specialization and multi-model orchestration. The discussion explores how switching costs are dropping, margins still matter in AI-native businesses, and why geographic location (particularly San Francisco and Tel Aviv) remains critical for tech founders.
Key takeaways
- • SaaS is oversold as a disruption target because AI has bigger opportunities: IT spend represents only 8-12% of enterprise spend, so companies will use AI to extend core advantages and optimize the remaining 90% rather than rebuild existing software.
- • Application layer companies will outperform foundation models by creating an aggregation layer that lets users access multiple specialized models based on use case, similar to how Cursor orchestrates different coding models for front-end and back-end work.
- • Margins matter differently in AI-native companies: while blended margins may be lower due to free trials and CAC-oriented spend, the durable margin profile of power users who pay 10x higher subscription rates (e.g., $200-300/month) justifies the investment.
- • Defensive moats still exist in a multi-model world—network effects remain powerful, and live proprietary data (health data, product-specific data) combined with commodity models beats cutting-edge models without access to proprietary datasets.
- • Founders should pick investors who actively help with introductions and lend their brand credibility, particularly at early stages, but ultimately the best founders know how to maximally leverage their VCs rather than operate entirely independently.
- • Founding authenticity matters more than polish: in enterprise, domain expertise and repeat founder experience are alpha, but in consumer, beginner's mind and willingness to be embarrassed about working on "embarrassing" products (companionship, creative tools) creates competitive advantage.
- • We are not in an AI bubble because supply allocation matches demand (100% of new inference capacity is spoken for), prices are rising rather than compressing, and subsidization is being intelligently directed by big tech labs to benefit consumers and startups.
Recommendations (6)
"YC is an awesome place to start an enterprise startup that sells to other enterprise startups"
Anish Acharya · ▶ 1:03:17
"Hard Things was the first honest business book. Most business books are full of shit and Hard Things is the first one that was authentic."
Anish Acharya · ▶ 1:10:26
"I'm a longtime transcendental meditation person. I've been meditating since I was a little kid, 25, 30 years. And it brings me this peace and joy"
Anish Acharya · ▶ 1:19:06
Mentioned (27)
More from these creators
OpenAI Kills Sora & Hits $100M ARR on Ads | Oura Going Public & Whoop Raises at $10BN
Why Margins Don't Matter for Early-Stage Startups | Gili Raanan
Why You Need a $1B Fund To Do Series A | SpaceX at $2TRN & Data Centers in Space | Groq's $20BN Deal
99% of Drone Companies Will Die & Why Anduril’s Products Aren’t an Ethics Debate | Matthew Steckman
Inside Figma's $1B ARR Machine | Shaunt Voskanian
NVIDIA Predicts $1TRN in Revenue: Everything You Need to Know From GTC & Anduril Lands $20B Contract