What The Fudge for May 17, 2026: Trillion-Dollar Valuations, Sovereign AI, and the Agentic Edge

It's time for What the Fudge, May 17th edition. I'm going to try to get it into 5 minutes one more time. If you didn't know, the top stories around the week included Anthropic getting valued at just under, I don't know, a trillion dollars. So that shows you that there's definitely going to be some interesting stories in the frontier AI movement. You've also now got the so-called customer forward deployed engineer or forward deployed engineer or customer forward deployed engineer. If you believe that is to be a new title of the future, that's something that now OpenAI is getting into. That's sort of a signal that they're going to have an enterprise consulting and possibly even cybersecurity business as they decide how to widen their aperture of what it is that they can do. You can probably expect other companies to do the same. There's also this concept around whether or not proprietary models or open weight models are going to be the next big thing. And so one thing that I think is interesting is, if you recall the whole Deepseek imbroglio that occurred previously, where someone said, oh, it took billions to do this thing, and then someone else comes out and figures out how to do it for millions, giving you similar effects, similar experience, I think it'll probably be more noise than actual concern. Think of it'll be sometimes the press gets ahead of itself. I'll just leave it at that. Then you've got the notion of the overall just general purpose, what happens when someone turns around and says, I need more compute, I need more infrastructure. And so some of these topsy-turvy kind of, are they enemies? Are they frenemies? Are they still competing? Partnerships around infrastructure taking place. And so as that happens, you'll see the, the SpaceXes, the Anthropics, the OpenAIs, the Googles, the Azures, and the AWSs of the world all sort of playing against each other, but at the same time playing with each other, and in the sense of where the infrastructure is coming from. Also some interesting threads on LinkedIn around where innovation is taking place. Is Europe truly going to compete with the United States. This, of course, conveniently omits the entire other side of a planet where Asia Pacific might be found. But, you know, it is the, it is the nature of the LinkedIn, Twitter sphere, I guess, X sphere now, people will share opinions online. But I think some interesting data points, I think if you look at the restrictions that are currently making their way through the courts around youth access to social media, There's some interesting stories around what happens if the so-called American AI moat was to somehow dry up, or in this case collapse. What would that mean? And then if you look at some of the interesting stories around, for lack of a better term, sovereign AI funds, which I've mentioned previously, as well as looking at how finance is actually going to be training up for more AI utilization, it's fair to say that the AI story is beginning to deepen itself not just within the organizational topology, but also probably within the team topology as it relates to more than just software development. And then if you think about what's happening right now with companies do— that do spin up their own standalone consulting business, that will most likely mean that the Fast or Big Five of consulting will have to respond in kind, because now you have technology companies literally getting into the consulting space. And so at this point, I hit the 3.5-minute mark, and so we With that in mind, just a quick roundup of things that I think are going to be interesting. If you do take a look at what's happening with the so-called agentic future, you're going to see some interesting elements. There's going to be the consumer side, which will be, you know, shopping agents that try to do dedicated shopping for you. I think there's also going to be some interesting stories around where and how these, uh, AI moat conversations form up, but at the end of the conversation around that AI moat, you would have to ask yourself, is it truly about only the large language models? And I still think these small language models, these fit-for-purpose computer vision at the edge, so-called kinetic AI, edge computing, is going to be the real breakout story. It won't always be about massive data centers somewhere in the country where you can get cheap power, cooling, plenty of space. So with that in mind, I'm almost at the 5-minute mark. I'm gonna wrap it up there. And no major updates on the NASA front this time. So if you want something about the space value chain, it's interesting that a couple of companies have been acquiring each other, but I'll cover that in a future update. Let's see how far they get. Thanks for listening to What the Fudge podcast.