MY FIRST MILLION · EXTRACTED
Brainstorming ChatGPT Business Ideas With A Billionaire
6 frameworks from a live AI brainstorm with a billionaire — on how to find the real opportunities in the AI wave before they're obvious.
Preview · 3 of 6 tactics
"Everyone is trying to build the AI. The real money is in building around the AI — the boring infrastructure, the specific workflows, the things that feel like too small an idea."
Sam and Shaan sit down with a billionaire operator and investor to brainstorm real businesses to build around AI. The session cuts through the hype and gets specific: where are the actual gaps? Who is being underserved? What will look obvious in 3 years that nobody is building today? The result is one of the most practical AI opportunity maps the show has produced.
Sell Workflows, Not Software
The billionaire's opening argument: most AI companies are selling tools. The real opportunity is selling workflows. A law firm doesn't want AI — they want their contract review process to take 2 hours instead of 2 days. A real estate company doesn't want AI — they want comps pulled and formatted in 10 minutes. The winner in each vertical won't be the best AI. It'll be the company that understands the workflow best and wraps the AI around it. 'The tool is a commodity. The workflow is the product.'
THE PLAY
Pick one specific professional workflow that currently takes hours of human time. Map every step. Then ask: where does AI eliminate steps? Build the workflow product, not the AI product.
The 'Last Mile' Problem Is Always the Opportunity
AI can do 80% of almost anything. The last 20% — the human judgment, the specific context, the quality check — is where most workflows fail. The businesses that figure out how to handle the last mile cleanly will win their categories. He calls these 'AI + human' businesses: the AI does the volume work, the human does the final judgment call. This is not a weakness. It's a moat. Pure AI can be replicated. Good AI + good process for the last 20% is much harder to copy.
THE PLAY
For any AI workflow you're building: explicitly design the last 20%. What's the human review step? Where does a human add judgment before the output goes to the customer? Build that process as carefully as the AI part.
The Boring Infrastructure Play
During the Gold Rush, the people who got rich selling picks and shovels weren't glamorous. The AI equivalent right now: prompt libraries for enterprise teams, AI output QA tools, fine-tuning services for specific industries, training data curation. Nobody talks about these at cocktail parties. But every company deploying AI needs them. He's personally investing in two 'picks and shovels' companies that most VCs overlook because the pitch isn't exciting enough.
THE PLAY
List the top 5 operational problems that companies trying to deploy AI actually face: bad outputs, hallucinations, prompt management, legal review of AI output, employee training. Which of those problems could you solve as a service business?
Subscribers Only
Unlock the Full Protocol
3 more tactics + Action Plan
TACTIC 04
The 'Reluctant Adopter' Market Is the Biggest One
TACTIC 05
Find the Job That AI Makes Unnecessary and Hire the Person Who Had It
TACTIC 06
The Subscription Beats the Project Every Time
Already subscribed? Log in
Newsletter
Get each new protocol the day it drops
One email per drop. No spam. Unsubscribe anytime.
MY FIRST MILLION · EXTRACTED BY PODEX