“There's an unspoken covenant that as a founder, you go down with the ship. For better or worse, it's changed a bit over the last year and I think it's disappointing, to be honest.” Enough said. This show is everything and more on: - What really happened behind the scenes - Did Google leave behind a treasure of an asset - How was the deal structured - How Cursor and Cognition deal with ever increasing reliance on Anthropic My 4 takeaways with @ScottWu46 👇

1. I Would Bet the Value of AI Goes Way Way Up The foundation labs (OpenAI, Anthropic, xAI, SSI…) are collectively worth >$500BN The application layer companies (Perplexity, Sierra, Decagon, Cognition, Harvey…) are collectively worth $50-100BN The value of these companies will go way up in the next five years. Love to hear your thoughts on this @btaylor @ttunguz

2. Sam Altman’s Bubble Theory In 2015, everyone was saying there was a tech bubble. He put out this bet with $100K on the line that in 5 years: 1. The top 6 Fortune 500 companies (Uber, Palantir, Airbnb, Dropbox, Pinterest, SpaceX) would be worth over $200BN 2. Mid-stage YC companies (Stripe, Zenefits, Instacart, Coinbase…) would 3x 3. YC’s 2015 Winter batch would be worth at least $3BN. I feel like we are in the same place with AI today. Which modern AI “bubble bet” would you make and why? @sama

3. What Matters to Win in AI Today Lots of compute, not lots of data. If you have a particular use case, having a small amount of data in that vertical in the correct environment will work better than large amounts of data. Quality over quantity. How does this compare to what you see in your models today? @amanrsanger @antonosika

4. Should Students Still Study CS Today in University A CS degree doesn’t just teach you how to code, it teaches you how to problem solve from first principles & how computer architectures work. It is a degree on how to think. That will always be valuable. Which core CS principle do you think is most overlooked by today’s AI founders? @richardsocher @sama @apartovi

Youtube 👉 Spotify 👉