How I Invest with David Weisburd · Sep 28, 2026 · 44m · 26m saved
With SC Moatti, Managing Partner at Mighty Capital and Board Chair at Products That Count
Mighty Capital plotted 550 companies by how defensible their competitive advantage is and by how much money they have raised, and only two landed on the line where more money meant more protection.
Most investors assume the moats of the software-as-a-service era carry over. SC Moatti said switching costs and proprietary data no longer protect an AI company, and that some of the most valuable companies have raised very little.
"Only OpenAI and Anthropic are on that diagonal and they are at the very top of the diagonal."
SC Moatti, Managing Partner at Mighty Capital and Board Chair at Products That Count, on How I Invest with David Weisburd. She was early at Facebook and late at Nokia, and her firm reads the conversations of a community of 600,000 product builders to pick investments. She said Mighty Capital has had six IPOs and six strategic acquisitions in eight years, including Netskope and Groq.
The full interview is covered here so you can skip it. 44 minutes of audio, 18 minutes of reading.
Here are the 15 arguments that matter.
Key Takeaways
Only 2 of 550 companies raised money in line with how defensible they are: OpenAI and Anthropic
Switching costs and data moats no longer protect a company in AI
Network effects and counter-positioning are the moats that work now
The IPO floor has moved to about $10B, and she said founders may need $30B to $50B of value before listing
The power-law venture strategy is a constraint for mega funds, not a rule for everyone
Most M&A happens between $100M and $1B
Chief product officers, not finance teams, now drive corporate acquisitions
Cutting jobs for a 20% efficiency gain is the worst way to adopt AI, because it draws the most resistance
Revenue per employee should rise by tripling revenue, not by cutting staff
AI agents break the old sales channel, because the user of the product is often an agent with no buying power
AI products still lack real personalization, the missing third of what makes a great product
Venture firms that only add AI note-taking are unlikely to survive
1. Only 2 of 550 on the Line
The host opened by asking what a moat means once AI is in the picture. Moatti said Mighty Capital mapped 550 companies in the seven powers framework, plotting each by the defensibility of its moat and by how much money it had raised. She expected the companies to sit along a diagonal, with more capital meaning more defensibility.
Only two companies sit on the diagonal, both at the top
"Only OpenAI and Anthropic are on that diagonal and they are at the very top of the diagonal." — SC Moatti
The other companies fell into two clusters.
Heavy fundraising has not bought defensibility
"Then there's a cluster of companies that have raised a ton of money, but aren't defensible, because they rely on moats that no longer work in AI, like switching costs, or even cornered resources like a data moat." — SC Moatti
The second cluster raised very little because the companies use AI to run lean, and she said they are very valuable. Their protection comes from two sources: network effects, where a product becomes more valuable as more people use it, and counter-positioning, the advantage a new entrant gets during a platform shift by doing the same job differently from the incumbents.
"It's really network effect and counter positioning." — SC Moatti
