Tech Concentration, AI Spending, Token Costs Add Fog to AI Outlook #shorts

The top 10 stocks are all AI related in the S&P 500. The top 10 stocks are about 40% of the S&P 500. So this is really what's driving the overall headline number and semiconductors are sort of at the tip of that sphere. The thing that people should be focused on, there are two things. If you're bullish, you say, "Well, Agentic AI really showed up on January 30th with Open Claw." And then you had Meta um introduce Muse and then that really took off. It's been number one on the free apps for Apple now for a couple of weeks and people are really getting excited about it. So that's the bullish side. The bearish side is if you look at the data and one of the things that people have focused on is is openweight models cutting and putting price pressure on what you can charge per token. And if you look at like the ramp index, which I think is pretty good, you can see that spending has been up about 9x from the beginning of the year through about August 2nd. But from August 2nd on, if you add OpenAI and Anthropic together, it looks like the spend for both of them has gone down about 11% over the last 2 months. So you look at that and you go, huh, maybe the amount of tokens being produced is not growing faster than the price pressure that we're seeing. Then you get this news that, you know what, the revenue figures are 50 billion, not 70. And you have to start thinking about it and you know forget today Micron's down from when they reported right beat and raise quarter. Samsung similar situation though they actually came in light. Um numbers were obviously spectacular but at the end of the day they were below street expectations. Those are all things that you have to work into this because these companies I think everybody who's been doing this a long time knows that they're over earning. You don't necessarily have multiples out of control, but what you do have is earnings out of control relative to what they are normally in the sense that you don't see 80% operating margins at memory companies. You don't see 80% gross margins even at these companies. And you have some investors who are hyper bullish thinking this is going to continue for the next several years. And I just don't think that's going to be the case. So you have all of these things coming together all at once, which is why I think you have to be pretty nimble. And that's why we've, you know, moved back and forth in terms of our thoughts based on what's going on with the AI trade over the course of the year.