Let me tell you about Leopold Aschenbrenner.
Columbia valedictorian. > Graduated at 19. > Got hired by OpenAI to help figure out how to control AI smarter than humans. > Then got fired for basically telling them their security was a national security disaster waiting to happen.
Most people would go lick their wounds. - Leopold wrote a 165-page manifesto.
๐ "Situational Awareness: The Decade Ahead" predicted AGI by 2027. It went so viral in tech circles that a UT Austin professor called it "one of the most extraordinary documents he'd ever read."
Then Leopold launched a hedge fund around the thesis.
In one year: $383M โ $5.5B.
So what's the actual bet?
Here's where it gets interesting for us as investors ๐
He's NOT loading up on AI chatbot companies.
His thesis: the real winners are the picks-and-shovels plays โ power plants, data centres, and the infrastructure quietly feeding the AI gold rush.
Think less "who's building the smartest AI" and more "who's building the roads AI travels on."
29 positions. One core idea. Massive conviction.
The links if you want to go deep:
Fair warning: the manifesto is 165 pages. I'd suggest starting with the 13F and working backwards ๐