Via 247wallst.com
Three of the most closely watched investors in finance independently loaded up on AI infrastructure plays in Q2, and the earnings data backing their thesis is hard to argue with.
When Stanley Druckenmiller, David Tepper, and Peter Thiel all move in the same direction at the same time, it tends to get people’s attention. Fresh 13F filings published in mid-August reveal that all three legendary investors made concentrated bets on AI-exposed mega-cap tech stocks during the second quarter of 2026, collectively deploying well over $1B into just two names: Amazon and Alphabet.
The positions #
Druckenmiller’s Duquesne Family Office opened a brand-new position in Alphabet, picking up roughly 336,300 shares worth approximately $120 million by the end of Q2.
On the Amazon side, Peter Thiel’s Thiel Macro reported a new $118 million stake in the e-commerce and cloud giant. That position now represents 28% of the fund’s roughly $419 million portfolio, making it the single largest holding. The move is particularly notable because Thiel Macro had reportedly sat out equity markets for several previous quarters.
Then there’s Tepper. Appaloosa Management didn’t initiate a new position in Amazon. It doubled an existing one. Tepper’s Amazon exposure now sits at around $900 million, accounting for approximately 15% of his portfolio.
Why these stocks, why now #
Amazon’s AWS division reported 37% revenue growth in Q2 2026. AWS’s backlog surged to $496 billion, suggesting the demand pipeline for AI-related cloud services isn’t just healthy but accelerating.
What’s interesting about the collective thesis here is what it isn’t. None of these investors piled into chipmakers or pure-play AI software companies. The bet is on infrastructure, the companies building and operating the data centers and cloud platforms that every AI application ultimately runs on.
What the smart money is telling the market #
Quarterly 13F filings are backward-looking by nature. They show where money was positioned at the end of a quarter, not necessarily where it is today. Positions could have been trimmed or expanded since June 30.
For the broader market, the filings reinforce what earnings data has been saying for several quarters: AI is showing up in revenue lines and backlog figures at the companies best positioned to monetize it. AWS’s 37% growth rate and $496 billion backlog aren’t projections or analyst estimates. They’re reported numbers. Disclosure: This article was edited by Editorial Team. For more information on how we create and review content, see our