According to CNBC, Dan Sundheim’s D1 Capital hedge fund initiated new positions in several AI-linked technology stocks during the third quarter, specifically buying into Nvidia, Meta Platforms, Broadcom, and Synopsys. The fund established positions worth more than $220 million in Broadcom, nearly $199 million in Meta, $115 million in Nvidia, and over $77 million in Synopsys. D1 also significantly increased its stakes in Applovin by 5.9% to holdings exceeding $601 million and boosted its Reddit position by roughly 5% to over $465 million. Meanwhile, the fund reduced exposure to financial stocks including Bank of America, cutting that position by more than 30% to $285 million. All these moves come as the AI trade faces a volatile November, with all four new positions showing month-to-date declines including Synopsys down 14% and Broadcom dropping more than 7%.
Betting big on AI infrastructure
Here’s the thing about Sundheim’s moves – he’s not just chasing the usual AI hype. Look at the specific names: Nvidia provides the chips, Broadcom makes the networking hardware, Synopsys creates the design software, and Meta needs all that computing power for their AI ambitions. This isn’t a scattergun approach – it’s a calculated bet on the entire AI infrastructure stack. And that $601 million position in Applovin? That’s betting on companies actually monetizing AI through mobile advertising. Basically, he’s covering the whole value chain from silicon to software to end-user applications.
The timing question
Now, the obvious question: why buy into these names right before they all start sliding? These stocks have had massive runs, and everyone’s talking about bubble territory. But professional investors like Sundheim often think in quarters, not weeks. He might be looking past the current volatility toward what happens when enterprise AI spending really accelerates in 2025. The recent pullback could actually be creating better entry points. Still, it takes guts to add positions when everyone’s getting nervous about valuations.
The financial sector retreat
What’s equally interesting is what he’s selling. Cutting Bank of America by 30% and reducing Capital One and Apollo positions suggests he’s rotating out of traditional financials and into tech. That’s a pretty clear sector bet. Financial stocks have been performing well with higher interest rates, so this isn’t about dumping losers – it’s about reallocating to what he sees as higher-growth opportunities. When you’re managing billions, these kinds of sector rotations can be more telling than the individual stock picks themselves.
The Reddit surprise
And then there’s Reddit. A social media platform becoming one of his top five holdings? That seems random until you remember they launched AI-powered advertising tools earlier this year. The stock’s up 18% year-to-date, outperforming the S&P 500. Maybe Sundheim sees Reddit’s treasure trove of conversational data as an underappreciated AI asset. Or maybe he thinks their new advertising technology will actually work. Either way, it’s a fascinating contrarian bet in a space dominated by the giants.
