AI Companies Are Trying to Run Uber's Playbook but Missing a Critical Ingredient
In this solo episode of Very True, Alex tackles one of the hottest and most scrutinized topics in tech today: the elusive business model of Artificial Intelligence.
Right now, almost every major AI company raising billions is running a familiar playbook. It’s a spectacular playbook mastered by Uber: raise an unimaginable amount of money, lose it on purpose to change consumer behavior, and turn the prices up once you own the market. But Alex points out that while everyone is quoting the Uber playbook, nobody is copying the exact piece that actually made it work. By holding Uber’s historical financials up against today's AI landscape, Alex breaks down the difference between a subsidy that is a strategic investment and a subsidy that is just a prayer, revealing why the destination for AI keeps moving. And what does it all depend on? Well-understood Unit Economics that match the business model.
Episode Highlights:
- [00:00] The Uber Playbook and the Missing Piece: Alex introduces the core strategy driving AI fundraising today and highlights the stark difference between an unoptimized business and an misunderstood business.
- [01:05] The Simplicity of Uber’s Unit Economics: Breaking down Uber on a single slide. Alex explains how marketplaces balance supply and demand at a local level, creating a knowable mathematical machine where Customer Acquisition Cost (CAC) could be recovered in five months before printing pure profit.
- [03:45] The 20-Billion-Dollar Land Grab: A look back at Uber’s aggressive fundraising and financial journey from its 2019 IPO ($13 billion in revenue at an $8.5 billion operating loss) to its massive 2025 turnaround ($52 billion in revenue and $9.8 billion in free cash flow) fueled by behavior change and a shift toward delivery.
- [08:15] AI's Reality Check - Subsidizing in the Dark: Why the Uber playbook is breaking down for AI labs. Alex discusses the lack of unit economic clarity, the challenge of defining an AI "unit" (is it a token, a query, or an agent?), crumbling switching costs for users, and why AI companies are subsidizing toward a destination they cannot yet draw.
- [13:10] Cooking Up a New Model: Alex shares how he is using his mechanical engineering background to build a predictive, pressure-tested CAD model for AI unit economics rather than just feeling through the darkness.
Links & Resources:
- Verissimo Ventures: https://verissimo.vc/
- Follow Alex on LinkedIn: https://www.linkedin.com/in/alex-oppenheimer/
- Uber Financials: https://docs.google.com/spreadsheets/d/19cNzPABWoB8YY8xRHA9UI6xyGdyOEFsZvXQUipwvKWg/edit?usp=sharing
- Unit Economics Overview: https://www.loom.com/share/aaddd4018b0a47cc80c6dd64e00846fc
- Cal Newport's AI Reality Check: https://open.spotify.com/episode/4eV36PZCnUxc3TxSybfVN3?si=75131e6cd91147d9
About Very True: Hosted by Alex, Very True by Verissimo Ventures explores the honest, unvarnished stories of founders and the real problems they are solving. We look past the hype to find the truth in technology and entrepreneurship.
Creators and Guests
