Just noticed something interesting about two popular AI stocks that Wall Street is actually pretty bearish on right now, even though they've had monster runs.

Palantir and Micron have been absolute darlings in the AI trade over the past year. Palantir nearly doubled, Micron more than quadrupled. But here's the thing - some serious analysts think both are seriously overcooked at current levels.

Let's talk Palantir first. The company's software is genuinely differentiated. They've built this ontology-based architecture that gets smarter as it processes more data, which is different from typical analytics tools that just do reporting and dashboards. Forrester and IDC both rank them as leaders in their space. Their financials back it up too - Q4 revenue jumped 70% to $1.4 billion and they hit a Rule of 40 score of 127%, which is basically unheard of in software.

But here's where it gets sketchy. The stock is trading at 209 times adjusted earnings. That's wild. Even for a company growing earnings at 57% annually through 2027, that's an insane multiple. Brent Thill at Jefferies has a $70 price target on it. Current price is around $157. That's 55% downside if he's right.

Then there's Micron. Memory chips are crucial for AI infrastructure - both DRAM and NAND flash. Micron's the third-largest player and they actually gained market share from Samsung last year. Q1 results looked incredible - revenue up 56% to $13.6 billion, earnings per share up 167%.

Here's the problem though. Those gains came from a supply shortage that drove up prices, not from any real competitive advantage. Memory chips are basically commoditized. William Kerwin at Morningstar gets this - he's set a $225 target on Micron shares, implying 40% downside from the current $380 level.

The memory chip business is cyclical. That shortage will eventually flip to oversupply, prices will crash, and Micron has no real moat to protect margins when that happens. Wall Street actually expects earnings to peak around fiscal 2027 then start falling hard through 2029.

So the paradox with both these AI stocks is pretty clear. Palantir has a real competitive story but the valuation is disconnected from reality. Micron's cheap on paper but that's because the cycle is peaking. Neither looks like a great entry point right now if you're thinking about risk-reward. If you're already holding either, maybe think about trimming positions rather than adding.