What The Fudge for March 15, 2026

Jay Cuthrell's first solo podcast episode covers AI tools like ChatGPT and Claude, open-source agents, product engineering trends, digital sovereignty, and space developments including Artemis II. He recorded it on Apple Voice Memos and plans to upgrade to a Shure microphone.
Chapters
Alright, we're gonna try it. Welcome to What the Fudge. I'm gonna do the following things. If, if you're not aware of my blog on fudge.org, I've been toying with the idea of doing a podcast. This is the first attempt at that. You gotta start somewhere, and if you're not embarrassed by your first release, you have waited too long. So just as a quick roundup, Everything these days seems to revolve around subjects like AI, but there's also this concept of sovereignty, whether that's digital sovereignty in the technology space, meaning that your technology is usable within some confine, whether that confine is to your country or locally to you, or maybe your own personal digital sovereignty. But there's always this concept of digital sovereignty coming up. Around what you should have access to. You know, for example, should you have access to personal AI or do you have to always go out to an external third-party service provider? And when you talk about AI, of course, there's the ones that get into the news. This is the concept of what's your default? Is it ChatGPT? Are you into Anthropic Claude? Are you into Google Gemini? Um, when I say ChatGPT, of course that would be OpenAI, but everyone's got their approach to whether they've decided on what's their daily driver. But the market is going to have different segmentations. There will be people that are developers. Those developers might feel strongly about, say, Anthropic Claude. There might be people that are more generalist and they just feel very strongly about what they've been doing with OpenAI's ChatGPT. There may be people that are, because they use Gmail, they started playing with Google Gemini. It all just depends. But then there's other things like open source agents, and the open source agents are getting more mainstream. So if you've heard of OpenClaw, that's this concept of, you know, a command line kind of thinking world going into, for lack of a better term, a success-based or set it and forget it approach to running an AI.. And in this case, OpenClaw has been, you know, thought of as, okay, well, what if you did that OpenClaw but for everything? And if you follow OpenClaw on GitHub, then you know it's pretty popular. It's, I think, one of the most favorited or starred projects on GitHub. But of course, if you think about this in the larger scheme of things and some of the folks that started AI back in the early days, Even some of those founders have moved on. So Yann LeCun, you'll hear his name a lot. He's actually now got this startup called Advanced Machines Intelligence, or AMI. They've raised, uh, well over a billion dollars, and they're going to be developing so-called world models that are more about the physical world than just the GPTs and text patterns, or creating, uh, three-toed sloths carrying a bucket of fried chicken that, uh, for whatever reason someone decided that was a great use of image creation capability. So it'll be interesting to see these things play out. But if you look at the future of where this is headed in terms of work, what it means to have access to data, you may have heard of or been a part of this progression from, you know, the age of various software engineering titles. You probably remember there was platform engineering and then there was product engineering. So now there's going to be product engineers that really think of this in the sense of true product management, but in the purest sense beyond what would have been a team of people now sort of embodying that empathy plus development capability inside of a new role. You may have also heard of things like the context layer, where in the past you may have been a prompt engineer, now you're more of a context engineer. But the thing that you're really hunting on there is What happens when software, which was supposedly eating the world, if you remember that Andreesen-Horowitz or Marc Andreesen phrase, but now there's Naval who is basically saying that it's AI that is actually eating the software. So whether or not you subscribe to those pithy phrases or not, it is interesting that AI is becoming part of all product engineering focuses. Every shipping software product is going to have to include some sort of an AI feature. So if you remember, I had also done the— in the past, I had the space value chain where I had forked the Landscape 2 project from the Cloud Native Foundation, or the CNCF. So I still think that's interesting. But if you're not familiar with some of the larger players, whether that's Axiom, for example, they're getting new funding. A lot of this is around research now, where we're seeing more interested in microgravity research, doing deployments into space to move the science into space where you do have these unique microgravity, so-called zero-gravity environments. And then you've got things like people wondering what's the latest and greatest with NASA, with Artemis II, what's going to be going on with those launch attempts as we, you know, go back, quote unquote, to the moon. And then looking at things like the European Space Agency, what they're doing. And some of the other spaceports that are coming up around the world. I think it's important to understand that most, most nations that have the ability to put something into space will be trying to put things into space. I don't think it's going to be necessarily one or two countries. I think it'll be every country that will eventually want to have some sort of spaceport or some way to put things into space. Much like the sovereignty we talked about before, just like many countries are getting for the first time, their own sovereign satellites up in space. And then last, rounding out some of the thoughts for the week, is looking at what it might mean for those that are thinking about what comes next, what I'm doing with this newsletter/podcast. I'm currently recording this using the Voice Memo app on my Apple laptop, so I don't know that there will be amazing call quality, but I will eventually break out the Shure microphone set, do it much like you would expect with a podcast where I'm kind of hunched over near a microphone talking into it. Whether or not I use video has yet to be decided, but I just wanted to get this recorded for tonight and then take this as, again, the first time. I don't know where the next one goes, but I'm actually curious to see how the transcription works out. And with that, thank you for listening to What the Fudge, the first edition, on Sunday, March 15th, 2026.
More from Jay

What The Fudge for September 20, 2026: Design Fiction, Autonomy, and Kinetic Infrastructure

What The Fudge for July 26, 2026: Infrastructure, Intrigue, Distillation Drama, and HWaaS

What The Fudge for July 19, 2026: Open-Source AI, Regulatory Headwinds, and Enterprise Capital

What The Fudge for July 12, 2026: On-Device AI, Memory Boom, and Big Tech Pivots

What The Fudge for July 05, 2026: AI Infrastructure Oops, Token Rationing, and Digital-Only Mandates

What The Fudge for June 21, 2026: Semiconductor Crunches, Export Playbooks, and AI Slop

What The Fudge for June 14, 2026: Trillionaires in Orbit and Agentic Intersections



