Just caught something interesting about how one of Wall Street's sharpest value investors is positioning right now. Bill Ackman's been making some notable moves with his AI exposure, and it's worth paying attention to.

So here's what happened: His hedge fund Pershing Square had been riding Alphabet pretty hard since early 2023 - we're talking 2.2 million Class A shares and 8.1 million Class C shares. But in Q4, they basically cut that position down dramatically. Sold off 86% of the Class A stake while barely trimming the Class C holding. Interesting move, right? Bill Ackman's known for being a value guy, so this rotation tells you something about how he's reading the market.

Where'd that capital go? Mostly into Amazon. Last year he started building a position there, added 5.8 million shares in Q2, and then doubled down in Q4 with another 3.8 million shares. So now Amazon's become the bigger AI play for him.

I get why he might've made this trade. Both Alphabet and Amazon are building out these vertically integrated AI ecosystems - they're both dominant in cloud (AWS at 28% market share, Google Cloud at 14%), both working on custom chips to reduce dependency on Nvidia, both exploring AI across consumer tech, autonomous vehicles, robotics. Pretty similar strategic positioning. But here's the thing: Alphabet had an absolutely insane run last year. Amazon? Relatively flat. So from a value perspective, rotating out of the expensive one into the cheaper one makes sense. That's classic Bill Ackman playbook.

But there's another move that caught my eye. In Q4, Pershing Square also initiated a fresh position in Meta - 2.7 million shares. This one's interesting because Meta might be the most underrated AI story among the mega-caps right now. Everyone's worried about their metaverse spending and whether they'll be disciplined with AI budgets, but the actual numbers are wild. Their Advantage+ product is running at something like $60 billion annual revenue, and it's basically automating digital advertising in ways competitors can't match yet. If they can scale that across 3.6 billion daily active users on Facebook, Instagram, and WhatsApp, you're looking at a pretty durable competitive moat.

What's interesting is the timing. Both Amazon and Meta have gotten hit recently - Amazon after announcing massive capex plans, Meta still in that 'prove it' phase where investors are skeptical about AI spending. But if you look at forward valuations, they're actually pretty attractive compared to where they've been since the AI boom started. Bill Ackman seems to be betting that the market's overweighting near-term uncertainty against long-term AI upside, and honestly, that's not crazy thinking. When institutional money managers like him are buying the dips, it's worth at least considering what they're seeing.