In this solo live stream, Rich Schefren discusses recent AI model behavior, noting that newer versions (Claude 5.6, Fable, Opus 5) have become more literal and less adept at following nuanced directions. He pivots to his core thesis that, just as attention became the scarcest online commodity in the mid‑2000s, human judgment is now the next limiting factor for competitive advantage. Schefren explains that judgment must be captured in real time, and he offers a practical solution: his free “imprint” on GitHub (accessible via Atlas) that records user feedback to embed personal judgment into AI interactions. He references his 2007 “Attention Age Doctrine,” his current manifesto (hosted at strategicprofits.com/manifesto), and the broader Agora ecosystem, arguing that AI currently lacks a point of view and that embedding human judgment will differentiate successful businesses. The stream ends with no audience questions, a promise to return the next day with a live build, and a brief sign‑off.
Chapters
Hello, hello, hello, it's Rich Schefren here. August 10th, Monday, and back in the saddle, uh, and we're talking about everything related to AI. And I'm just getting set up here to be live on Instagram as well. There we go. And, uh, if you're here and you're watching me please by all means say hello so that I know who's here. You know, it's interesting, I was just talking to my daughter, my oldest daughter, Ava, and she was telling me that her Claude is not following directions as well as it was just like a month or two ago. And I wonder how many of you have had a similar experience. Good morning, big. I think quite a few people have, and I think that's primarily because in the step up to new models, the— I find that they're not as good at following. So they're much more literal. And if you haven't found that, I'm kind of curious as to why. But I'm curious what your experiences have been with the new models, whether it's 5.6 or whether that's Fable or Opus 5. Um, I found them to be incredibly powerful, but I have had some challenges with them. I think it's because they're a little bit more literal. Um, but I'm curious what you guys think. And in addition to that, what I've been working on is also, um, getting my StreamYard up and running. And, uh, I've been getting more and more feedback about the manifesto. If you haven't read it yet, I strongly recommend that you do. It was a labor of love and it's really about what I see as really the next competitive advantage that you'll find online. Um, you know, back in 2007, I wrote a report called The Attention Age Doctrine 1 and 2, and it was talking about how attention was going to become the scarcest commodity online. And, you know, before social media and before infinite scrolling and all those things, uh, it wasn't so obvious as it is today. The next big, um, sort a scarce commodity, the next big constraint is going to be judgment, and, uh, just like attention was. And so you are either doing things right now to capture your judgment, or you are right now wasting your judgment. It's one or the other. And the reason I say that is, is that judgment is not something that you can go backwards and capture. It's got to be captured in real time. And because it has to be captured in real time, the best time to start was 20 years ago. The second best time to start is today. And so if you want to start capturing your judgment in AI, it's really pretty straightforward. Um, all you have to do is tell your AI to go to, um, GitHub and do a search for Rich Schefren's imprint. And you can download it for free. And what it will do is it will start capturing your feedback. Every time you give your feedback to AI, it will capture it. Hey, Nicole, good to see you. I hope we see you down here at some time soon. And so that's what you need to do for your imprint. Just go to Tell Your AI, that Rich Schefren's imprint on GitHub, and If you can't seem to find it, you can just type Rich Schefren imprint GitHub in Google. Um, it might not come up, but if it doesn't, you can always just click on what does come up, which will be Atlas, then click on my name, Rich Schefren, and you'll see the imprint. And the imprint is what will begin capturing your judgment. And when I was, uh, talking on Friday about what we were doing at Agora, that's a big part of it. Um, capturing, uh, judgment. Because it really is like the last bastion of capitalism. I believe, uh, that it really will be the one thing that separates. And if you think about it like this, right? Think about the most successful people on the planet. They're not the smartest people on the planet. They have the best judgment. Judgment is everything at the end of the day. And right now, AI has no point of view. Because it has every point of view. And because it has every point of view, it has no point of view. That's why it's so quick to bend on anything that's not factual, because it doesn't have a point of view. You have to provide that point of view. And when everything else is the same, that point of view is going to be the one thing that differentiates you from anyone else. And you either will have that built when it's needed or you won't. And so that's why I wrote the manifesto. I wrote the manifesto all about what it takes to work today, like to make money with AI today, right now, and how to know whether you're building a competitive advantage for the long term, which would be to capture your judgment, and what to do if you're not even close because you're getting in your way. And that would be, uh, Wave 3. And you can read it all at strategicprofits.com/manifesto. And you can get the imprint at GitHub and you can also get Atlas at GitHub. So anybody have any questions? Because I would love to answer questions, but got a bunch of you watching on all these different channels. But nonetheless, I don't see any questions coming in. Some good mornings, but no questions. Questions. And if there's no questions, then maybe today will be a very short episode. I think it will be. All right, well, I'm going to jet, uh, there's not too many people on, nobody has any questions. I guess I will start again tomorrow, and actually I'll bring something to build in case, uh, nobody has any questions tomorrow, and I will just build relentlessly until someone does. Alright, so have a wonderful week, and I will see everyone next week. I mean, I'll see everyone tomorrow at noon again. Till then, adiós.
