Does AI Make Us Dumber?

Welcome to another episode of the Knot Brothers podcast. Today is episode 10 and we'll be talking about whether AI in general makes us as a human species dumber, smarter, more intelligent. Let's debate. ⁓ We were talking pre-show Ryan about in higher ed in particular, there's this potential trend of moving away from technology based ⁓ writing assignments. And instead forcing students to turn things in on in like blue books that are written in class or with a monitor of some kind because cheating is so rampant. Even pre-AI, I think that was probably very, very true. But now in an AI era, it's, ⁓ it is incredibly easy to say, please write me a three page paper about ⁓ intersubject. And it just spit out something that looks good enough to turn it in.

Yeah, I think there's...

There's probably a lot to unpack in that too. We talked last time about what constitutes work. ⁓ Often when people create something or ⁓ rather have AI create something or some combination thereof, they'll bring it back and present it in a professional setting as I created this thing. ⁓ Look at this thing I did. And that's generally accepted by ⁓ most everyone. ⁓ whether AI created it or not. So the question is like, ⁓ should it be any different in an education setting? I think your goals in an education setting are somewhat different, but there's also a lot of education that is just ⁓ busy work for the sake of busy work. So like, when we think about things like writing five page papers, I don't think that writing five page papers are particularly, I personally don't think they're particularly helpful anyways. I think they're. time-filling activities that don't really provide as much benefit as ⁓ Others might think so ⁓ I would argue the like the premise of the thing in the first place but like the going if we accept that as a thing If I prompt if I write it by hand or I prompt an AI to write what I think is a good paper Should it matter?

think it depends on your... ⁓

I think it does, depending on your goal.

And that's the, that's the, well, I think that's the question of like, does it make us dumber? And I think that, I mean, we could some, we can make the podcast like three minutes long, right? Like does AI potentially make humanity dumber? Yes. End of podcast. Right. Like ⁓ I think the, ⁓ The potential for that is incredibly incredibly high because what you're doing is you're effectively delegating your brain ⁓ To ⁓ something else not even someone else ⁓ And at that point You're not learning any of the subject matter, right I think the the ⁓ Intent probably behind like write a five page paper about something is so that you can Articulate that you understand the topic that you understand what's going on You can describe it well all those my beef with the five page paper is like I'm a pretty brief person direct to the point person so ⁓ You know, I can't talk about anything for five pages worth of stuff, but I can give you like three really solid ones so That's when you just start messing with the spacing and figure out what you can get away with.

Yeah, I there's there's no doubt where we landed last time. And I think it's a good place to to land in terms of measurement is what are you actually trying to measure? What what do we care about? Do we care about memorization? Do we care about sentence formation, grammar, spelling? Do we what what are we trying to measure? And I think in the context of what we're talking about here and ⁓ cheating being rampant and writing papers, I think it's critical thinking and taking that critical thinking and putting it into a persuasive argument in written form. That's what's trying to be measured. ⁓ And ⁓ if you can't do that in an articulate way that makes sense, ⁓ throw grammar and sentence structure and that kind of stuff aside, although those are tertiary things that you would measure, it's are you making a persuasive critical thinking oriented argument ⁓ associated with this? Or did you prompt someone with a question or something with a question and it came back with all the critical thinking to your point, you outsourced your brain.

I don't know. I think that ⁓ I agree that, you know, it depends on what we're trying to measure. And the question is like, ⁓ our measurements flawed? Is there a different way that we should be? Does AI push us to ⁓ a different paradigm of education and not to make all this like education focused, right? But ⁓ it's obviously an easy one. The last. time we spent a lot of time talking about ⁓ education and AI involvement in education, but this also, you know, we talk about the intelligence level ⁓ of people and how that's being impacted by ⁓ AI and obviously brings us there. ⁓ I think you know much in the way that like ⁓ I don't really need to know how to ⁓ perform complex calculations because I have a calculator to do that and I always will and ⁓ that's just not something that's valuable to me. ⁓ I think that becomes a question that we have to measure things through too is like, what are we, ⁓ what's valuable and what do we consider valuable? Is it something that is, ⁓ is just for the sake of learning and for like.

muscle kind of expansion and contraction and during like the education process ⁓ or is it real world skills? Because I think the two are at odds with each other. ⁓ Right now, ⁓ if we're training for real world skills, we shouldn't be demonizing use of AI. We'd better be teaching you how to use the shit because employers are ⁓ like, they're not preferring it. They're demanding it. So if you ⁓ went to school when you were a kid that came out of school and you went to ⁓ some radically different education system that didn't use computers and everything is written and whatever, right? To some people that sounds fantastic, right? You're gonna come out and you're just gonna be so much smarter than everybody else. You're not, you're gonna be incredibly behind because you're completely ignorant of all of the technology that all of your peers already have experience with. And you'll be passed over for every job, you'll be perpetually behind.

I think the question that you really need to ask and really what you're talking about is the difference between ⁓ resourcefulness and intelligence. Resourcefulness is using the tools that you have at your disposal in the most opportune way possible. And that's incredibly difficult to measure. It's why I've forever hated standardized tests because they're not even really a measurement of intelligence, they're really more of a measurement of memorization. And how you apply that memorization to whatever the formulaic answers are. Right. That's why AI is really great at taking them because it's very formulaic in nature. Whereas resourcefulness is. Right. It's in a sheets. So resourcefulness and grit are two things that are incredibly difficult to measure. How much harder am I willing to work than this other person at being resourceful? I may not be as intelligent, but I may have those other two attributes that are far exceeding that person. And so therefore I'm a much better candidate for whatever this is. ⁓ so.

and ⁓ it cheats.

In the context of what we're talking about though, intelligence ⁓ is ⁓ different than memorization. ⁓ It's about critical thinking. ⁓ Can I put together in my mind the rational thought process and thinking structure not to get the direct answer, but to understand the logical steps that it takes to get that answer? ⁓ so the steps to get the answer... could involve various tool sets. could involve Google searches. It could involve freaking encyclopedias. It could involve AI. It could involve any other thing. But going through that logical progression of how do I get to the output that I'm trying to get in the best way possible is very different than just prompting your way through that loop.

Maybe. I think everybody approaches it differently. ⁓ I think that even when we look at my use of AI versus your use of AI, they're completely different. The way that we interact with it is completely different. Even if we're ⁓ seeking the same goal or even if our outputs look almost exactly the same, the pathway we took to get there is completely different. And it's, you know, partially how our brains work and how we're wired and ⁓ just the culmination of our past experiences with them.

I had another thought and I completely forgot it. But I think the... ⁓

I don't know. ⁓ It's a really, really tough thing because I get theoretically what we're trying to do and the fear that we have around AI is kind of this infinite access ⁓ encyclopedia that will make shit up. at your fingertips, right? ⁓ And ⁓ a large part of education and education systems and what we value, what we consider to be intelligence really, is memorization, right? The smartest people. ⁓ are generally considered the people who have the most knowledge, not necessarily the people who are the best at getting things done or the most efficient or ⁓ whatever. That could be a totally different thing.

Mm-hmm.

So there's an argument to be made that like, you know, ⁓ in some ways AI will and is going to make us dumber. And in some ways, maybe that's okay. Because it removes things that we don't need to, we don't need to know anyways.

I agree with that in some hand. And I think there's this other argument that's, ⁓ it's very related to what we're talking about, but it's completely different, which is this idea that learning should be hard. It should require work. It should require some sort of sacrifice. ⁓ It should require some sort of, it's the thousand hour rule, right? How do you become an expert at something? You have to do it over and over and over again. You have to know that there's a path that sucks and the path that sucks makes you, makes you great. at whatever it is. so learning that grit, learning that, that process of doing hard things, when things become so easy, that things that used to be hard are now literally a prompt away. We remove that learning that is not just the subject matter. It's the life skill of things are hard and you need to learn resilience. You need to learn how to push. need to learn how to fail, you know, learn how to get better. Those are things that don't come naturally. ⁓ and ⁓ younger generations are definitely shying away from hard things in that way and hiding behind screens. You see it in relationships. You see it all over the place. And now AI is just the one more ⁓ step in that same direction of ⁓ being opposed to doing hard things in real life.

⁓ Maybe. ⁓ think that...

I think the idea that everything has to be hard is a bit of a fallacy, Certain things don't need to be. ⁓ Obtaining food doesn't need to be difficult anymore. ⁓

I didn't say that.

I mean, that's just a controversial opinion though, right? Like, ⁓ so there are certain things that obviously you don't want to go back to. But the fact is we've replaced those hard things with different hard things ⁓ to give the same life lesson. So the question becomes, what's the next hard thing to teach the same life lesson that working on a farm used to teach?

Well, that's where I was going is, are we just at another one of those inflection points and we don't know what to do with it? Where, you know, when we, when we sort of evolved from ⁓ farmers to sort of more of the information age that we obviously there were steps in between, but ⁓ you know, it used to be you woke up in the morning and you worked on the farm and that's how you got food. And that's like, that's how you did everything. ⁓ I've never worked on a farm.

Probably. Yeah.

I've never killed my own food, right? Like, ⁓ yeah, it's a whole different story. ⁓

My wife's trying to make my house into a farm. ⁓

You know, it's not, it's no longer, ⁓ that's no longer part of what we have to do. And that was considered normal for a long period of time. So I think if we rewind the clock to those people, they would say, well, you guys are all just lazy and you should just grow your own food and do your own things. What it did is it freed up cycles right now, because I don't have to worry about tending to the crops and ⁓ plowing the fields and all that. We're able to build computers and software and, you know, explore ⁓ avenues that didn't exist back then and not only did they not exist, but you wouldn't have the mental cycles or the time to invest in those activities. So is there a possibility that what we're talking about now, what has been sort of the measure of intelligence and success and all those things for the past 20 years is now fundamentally useless or worth less than it used to be? And now there's some new ⁓ something. ⁓

So you're.

that we haven't really figured out yet, that's going to take its place, that now becomes what we invest. It's the human element of all of those things, and certain things just get delegated to AI, or it's considered fine to use AI to replace it that matter.

There's no question that we're in a paradigm shift. And so, ⁓ you know, the biggest questions about the paradigm shift are what are humans going to do when ⁓ AI is capable of, you know, in the world of an utopia society where humans don't have to do anything at all. That's the biggest question. ⁓ And I don't know that we're anywhere close to being ⁓ that or that it'll ever be a thing, but it is a worthwhile endeavor to think about like, okay, well, how do you ⁓ How do you have human purpose in a world where you can do whatever you want to do, but ⁓ for ⁓ the entirety of human existence, we were made to work. And that work is what gave us in a lot of cases purpose. And if it wasn't work, then it was creating something, which is a replacement of work. So it's. ⁓

Well, I think that's very like American and sort of fucked up view that we have anyways is like, most people in America, you live like your life is work. And like, no different, right? The majority of my life is spent doing work building business is always has been. And that's just a part. But other people in other countries think we're insane. ⁓ And we think they're insane for not right? It's the ⁓ it's a it's a weird like positional difference, but ⁓

I mean, work is a relative term, right? So if everything is so easy, then what do humans do in a society of ⁓ massive abundance? ⁓ I don't know. It's not about like going to work for a paycheck necessarily. What do we do?

and

I've got no shortage of hobbies. ⁓ And like maybe that's the answer, right? Like today we do work ⁓ and for years, right? We've gone to work doing things ⁓ and for the most part we've enjoyed and wanted to do. ⁓ Look, we've had plenty of clients and plenty of projects that... ⁓

⁓ Hahaha.

I would rather not wake up and work on those. And luckily, I mean, I'm fortunate enough to say that we don't have any those today, but certainly have, right? So, you know, if Elon's world, right, like what he's claimed is we'll have universal high income, which is just universal basic income, just with a bigger number. So everybody will just have whatever they need and you'll have, like, everybody has access to everything. So what would I do, right? If I woke up tomorrow and was, you we have universal high income and I can no longer achieve, I can no longer work to achieve more or anything, right? Like it's all just level set. Well, then I just go work on the things that I want to work on for fun. I tinker with computers and... continue building on Omarchi. ⁓ Omarchi is a fascinating one, actually. That's a good example of what would I do. ⁓ We get the question all the time. ⁓ I've gotten the question a bunch. I've seen people lob that question at David on ⁓ X a ton. ⁓ How does this make money? It's always the first question. When I tell people about the project itself, they're like, well, how do you make money with that? And the whole time, the answer is we don't. The intent is not to make money. We're not building this because it's profitable or because there's some commercial avenue or building it because there's a genuine interest in computing. It's really fun and ⁓ it needs to exist. So we're just going to make it exist. ⁓ And I think you would have a continuation of that.

And I think for... think for a certain subset of people, think that's true. ⁓ think there's ⁓ another subset of people that, I it's like the person that retires early, right? And they don't have a backup plan. That purpose, there's no real backup plan that's been thought about. You've thought about this because we've thought about what would happen if we ever sold businesses. And so you've put at least some level of thought into it. But imagine waking up with no backup plan. and no real, I don't know what to do today. I literally have too much time on my hands and I do not have a backup plan of what to do to fill my time with. Like that's a...

I think that's the problem and maybe that's what education should focus on is teaching people how to grow themselves by way of hobbies and other things. But I think that, again, going back to that sort of like fucked up mentality that we have, your life is work. There was a guy who lived across the street from me that I remember he retired. was like 62, 63 something. And I forget exactly what he had done. He was an engineer of some kind, I think. I talked to him. He's like, oh yeah, just I just retired. was like, well congratulations, you know, how are you enjoying retirement? It was like I hate it. I'm And the answer was for probably 40 years, this guy woke up in the morning and went to an office and did work for other people and then came home and that's what gave him purpose and that was his whole being. Everything was that and what he never developed in his free time or otherwise Was something what that was his right? He never developed a hobby or an interest or something outside of work ⁓ So when work went away when work was no longer necessary because he has achieved a certain age has achieved a certain financial ⁓ State of being there's nothing there ⁓ He didn't have ⁓ a golf hobby or a computing hobby or a project that he's working on or an interest in boats or like anything. And I think the guy actually wound up dying ⁓ a couple years later, which is a very common occurrence amongst people who retire with nothing to do. And then they basically like, it's like they get bored to death, literally.

super comment.

Yep. Yep. My dad says you sit, die. And that's kind of how it works. Right? I mean, the, you, if you don't have something to fill that time and we're, know, to kind of circle this back to what we're talking about the, in, the age of AI, question is, what do you do with the time that you have been afforded by AI being ⁓ incredibly efficient at something that you used to spend a whole week doing and it gets it done in an hour. So what do do when that gets extrapolated to a point where ⁓ obviously your ⁓ boss or your work or your business or whatever is going to demand increased productivity to fill that time slot. But what happens when that gets filled? And then that gets filled and then that gets filled. And so now we're extrapolating to a place where human ⁓ purpose is in question, at least as it exists today.

⁓ I think that's an extrapolation also that's like this isn't a straight line, right? I think we're My hypothesis with zero proof at all. So this is just a hundred percent my gut feeling and observation But I can't explain right. It's a hundred. It's a ⁓ human element in my brain, right? My brain sees patterns in information and is interpreting it some way It tells me we're at a bit of a plateau And I think we've been there for probably about a year where the models and the capabilities of AI. ⁓ have ⁓ sort of tapered off, right? We were on this massive hockey stick trajectory. ⁓ And if you extrapolate that trajectory out, which is what ⁓ people like Sam Altman and Dario want you to do, because that's what's good for their valuations, if you keep extrapolating that, you get into this world where you're like, well, there's nothing left for us to do, right? AI is going to be able to do everything. It's going to be able to solve every problem. It's going to be able to replace humans in every instance and do it faster and better and cheaper. we're already seeing cheaper be tested, right? There are instances in companies ⁓ where they're turning back to people in human capital because they figured out that using AI to perform certain tasks is actually not cheaper than people ⁓ and not better, right? So I'm paying more for a shittier result. ⁓ I think we'll see that continue to happen as these subsidy, like ⁓ we're all. drunk on these subsidized subscriptions and ways of using AI that are like not sustainable. So as those dry up, everything becomes more expensive and the usage will taper. But also the models themselves are not getting better. They're getting incrementally better, but they're not getting monumentally better because they've already absorbed the whole of the internet.

And not only that, people are creating things with them that are shitty, and they're absorbing that too. So now ⁓ you have this risk of like model collapse and model degradation that exists where, know, ⁓ Anthropic has mythos right now, right? Where they're like ⁓ mythos is so good that we can't release it because it would be irresponsible ⁓ and ⁓ it's just too dangerous. Come to find out.

GPT's supposed to have their version too now that they'd have released apparently.

Yeah, GPT 5.5. GPT 5.5 came out, and then when researchers who have access to both started head-to-head testing them, they're like, it's the same fucking thing. So, you know, I mean, we've seen this before, right? GPT 3 was too dangerous to release as well, and it's laughable now that we thought that that was good.

Good use of AI and critical thinking is a story I heard recently about IKEA. So IKEA was tempted to follow suit with a lot of these other organizations that were basically axing their customer service reps. What they actually did is they used AI to analyze all their customer service interactions and to really figure it out. And rather than ⁓ axing thousands and thousands of jobs, what they discovered is that those people, a tertiary piece of their job that they weren't even asked to do, it wasn't even on like their You know, it was like in the other duties as assigned list, but not really even written is actually be a concierge and a designer. So to actually help people find product, to help people put together different products, to help people with room design. And that was a whole, whole service line that they were completely missing. And so what they did was they invested in retraining those entire customer service groups to, to become ⁓ makeshift designers and to use AI to help them. along the way to help customers that are inquiring about these things and their sales have increased because of it. And their overall margins have increased because of it. I don't have the exact numbers, so I can't quote those, but ⁓ that's the genesis of the story is like, okay, well, that's a great use of critical thinking and AI. How do we evaluate some data sets that we already have and find a new application for something that didn't exist before?

So what's interesting about that, right, is I agree with you. I think that's a great case study for ⁓ how to use that. That would be discouraged in education.

You're not allowed to do that.

No, I don't think so. ⁓ I think it's different, right? So going back to some examples we talked about before we hit record. So what's an acceptable use of AI in education? Well, I've got notes, ⁓ I've got the ⁓ textbook, I have every piece of information that has been released to me. I can feed that to AI and say, help me study. Help me figure out how to study for this test. Quiz me, do multiple choice. You can get super creative about how you would do that. And that's a wonderful use of AI. What is not a great use of AI is say, take that same information and write me a three page paper. That's laziness, in my opinion. You're avoiding ⁓ the critical thinking part of what you're being asked to do as part of the assignment. And I guarantee you, as expressly said, what, prompting an AI?

Yeah, but that's not the hard part.

No, most of the time the summarization output is not the hard part. Writing a paper is not hard. Figuring out the shit to put in the paper, that's what's hard.

It's the research side is what you're saying.

Yeah, figuring out, doing the research, making the arguments, ⁓ figuring out how to articulate whatever it is that you're trying to articulate, that's the hard part. Writing the words ⁓ and summarizing all of that is hard.

⁓ But ⁓ we have the ability to literally take pictures of whatever we want in terms of textbooks or whatever, or just give ⁓ an AI a three sentence prompt, and it's gonna do all the things you just said for you instead of you doing the hard part, which is what you just said was the research side, right? And I agree with you, that is the hard part, the formulation of what do I wanna talk about, why? ⁓ Reasoning through.

Well...

different arguments to say, this argument right or is this argument right? And so that critical thinking part, it gets compressed into something that is a summary as opposed to the actual real documentation that you would normally go through and sift through yourself and use the meat computer to decipher truth from fiction. I'm not saying that AI shouldn't be used for those sorts of things. It's a wonderful research tool by itself. ⁓ But there's that fine line is a gray area. keep talking about is ⁓ where where it is, especially in education. Where does that? ⁓ Where's the learning part happen? Yeah.

Where does it? Well, at what point is it mine versus its? ⁓ Right? So like we believe that I could prompt AI to write a ⁓ three page paper about a topic and it will do it and it'll do it well. I think that's the fear. The fear is not that it'll do it. The fear is that it will do it well. And we will assume. I don't think it will. I I think with, if you were to just ⁓ give a basic prompt, it's gonna produce something, it's gonna be dog shit. It's gonna be a D student at best. Which if that's what you're good with, fine, go for it. You'll be a D student. I think that if you continue ⁓ to refine and. and massage on it and whatever at what point does it become mine and at what point does it become good I spent plenty of time with with AI trying to get it to write things that you know from a post to a You blog post or whatever is horrible It is horrible with incredibly detailed instructions of how to write and what the thing is or it's horrible They're still horrible. The best model today is horrible So if you turned it, I guarantee if we wrote a paper about anything and just said like AI write a paper and you graded it, it'd be like a D, maybe a C. It's gonna have shit that's wrong. It's gonna make no sense. ⁓ It's gonna be overly wordy for no fucking reason. And that's just what they do. And I think that... we've done the opposite right if we've if we've taken some of that and compressed it and pushed it and pulled it and and Worked my arguments into it and done had it do research and summarize a topic and help me put that in there. I Think that's fine. ⁓ I think that's the hard part

I think from a work output perspective, you're thinking about like, you know, if you're thinking about what we would care about as business owners, which is really output that isn't under some sort of copyright infringement, ⁓ right, which is the equivalent of plagiarism in ⁓ school, that's what we would care about. Like from that perspective, you're 100 % right. From that building blocks perspective that we even talked about last time of like, can you take a 10 year old and have a 10 year old write a college grade paper? Hmm. With AI get closer. You could, you could have it. Yeah. You could write something and a 10 year old that's good at prompting could probably be, could, could produce a pretty convincing college level paper. ⁓ that's not the point, right? The point is to try to teach that 10 year old the basic principles that they'll be using for the rest of their life on how to critically think, how to apply knowledge that is base level and our literal building blocks to, ⁓ to then apply to future works.

Sure, maybe write something.

But if it ⁓

by using whatever tools are available to you. And if you skip those building block steps and you're just jumping to the most powerful power tools that you have, you forget how to use a hand screwdriver in certain situations. And you forget how to use, ⁓ you know,

⁓ If I never need a hand screwdriver who gives a shit. ⁓ I'll give you you won't

But you will. ⁓ Like that's what it's a ubiquitous tool that every person in life will use, you know, in a modern society, you will have need for a hand screwdriver in the same way that you have need for multiplication. You'll have a need for multiplication.

Not if I have the Tesla robot. It doesn't. ⁓ Maybe. ⁓ I think, you know, when I was in school, one of the things that I had ⁓ an issue with, ⁓ because I'll never forget it, it was in seventh grade, ⁓ was handwriting. My handwriting is awful. It's still awful to this day because I refuse to fix it. ⁓ And at that time, my teachers approached me and they said, hey, your handwriting is awful and you need to fix it.

Okay.

And they gave me all of these reasons that it's going to affect my employment and it's going to all of my life because I'm going to have to write things for the rest of my life. And it's super important and valuable. My response as a seventh grader was, I don't care about that. I'll just type it. I type faster than I can write anyways. Because at that time I was already on the internet, had a computer dabbling with programming like 12. ⁓ And that was preposterous to them. They're like, what are you going to do, carry a computer with you everywhere? Well, turns out ⁓ we do. ⁓ But ⁓ that point in time was an inflection point. That point in time was where we were moving from handwriting and hand tools to ⁓ these new power tools in personal computers, laptops, and eventually cell phones that we had no idea existed. At that point in time, I forget what year that would have been. ⁓ But very few people had personal computers in their house. Almost nobody had high speed internet. I did. I was one of the first people that had DSL that was like 10 megabits per second. And that was like blazing fast. ⁓ So the idea ⁓ of a laptop wasn't something that most people could even fathom. The idea of a cell phone, right? A fucking computer you can put in your pocket. wasn't a thought. ⁓ Now looking at that, has that ever impacted me? Has not fixing my hand, has having completely horrid handwriting and not being able to write faster than I can type ever affected me ever? No. Because I moved past it, right? We as a society evolved past that being a need and being something that's valuable, so that it's fine. The question is which of the things that we do today are gonna have that same thing. There's gonna be something that we're talking about right now that we look back on and we're like, it's laughable that we even ever thought that that was gonna be something we continued to do. I don't know what it is. ⁓

I think.

⁓ You get no disagreement from me on 90 % of what you said. think, and I'm thinking about this from a dad's perspective, right? Like there are things that are part of handwriting development that actually formulate parts of your brain. So it's not actually about the handwriting, it's actually about doing something that's difficult for you that formulates different parts of your brain. Like I only know this because I have two kids that struggle with handwriting and they've gone to handwriting. helper people and they've explained the science to like why handwriting is important at all. So you're right, as a society it doesn't matter, but those micro-expressions that you're able to do with your physical hands have an impact on your brain development. And so for kids that are certain ages...

Are there ⁓ better ways to do that or different ways? like, are video games something video games have been proven to improve hand-eye coordination? So like, is it possible that rather than focusing on handwriting, actually, we should be focusing on teaching people to play first person shooter games because it increases all of those same muscles and flexes the same muscles just in a better or different way.

I have no science to confirm or deny anything that you just said other than to say those micro movements are important. And so how you have, how they happen is debatable. ⁓ but ⁓ those things, so, but you're right. Like the, basic premise here is like, it's the question of what are we doing today that will be proven irrelevant in the future. And I think that's the problem that we have with AI. We, you know, we experienced this paradigm shift a couple of times in our, in our lifetimes already, but this is the, this one has happened faster, probably 10 X faster. than the rest in terms of overall adoption and having to figure out how to react as a society. And at every level from a work level, from a higher education level, from a, you know, a high school, junior high, grammar school, all the way down to like little kids on screens. Like it's, it's, it's a, it's a huge, huge, huge impact on society. AI being the most recent thing that we have to think about and evolve past.

⁓ It has the distribution, right? ⁓ The last one that I think was probably as big was the invention of the internet. But the thing with the internet was while it did fundamentally change how everything works, right? It displaced ⁓ newspapers and magazines and ⁓ print advertising and print, you know, all of these things that we can think back to that used to exist in abundance ⁓ or were just normal that like barely exist today and in very different ways. And in part because those same thing like newspapers as a for instance, the physical newspaper does exist, it's barely, but wouldn't exist if those same organizations hadn't been able to react and basically create an online equivalent and sort of adapt with that. But they didn't have the distribution. When the internet came around, nobody had computers, nobody had cell phones. So it took a long time for the infrastructure to exist to even provide everybody access to it. AI had this just this circuit they could plug right into where everybody globally already has, we already have internet, we already have computers, we already have phones. So now you can distribute this idea so much faster and get in everybody's hands. What I think is what is kind of freaking everybody the fuck out about it, where whether it's, you know, businesses or personal, you just don't know. And especially when you start building, you know, new things like the, I mean, the rate at which you're able to build or create completely fake content, whether it's like fabricated images or fabricated video or just written content, like it's insane now, right? There's... There's so much more ⁓ that we have to worry about and it's continuing to evolve every day and we're not 100 % sure what to do about it just yet. I don't think that we ever will be completely sure what to do about it, frankly.

Yeah, ⁓ it's something that, ⁓ there's no question it's gonna stay with us. ⁓ Society has proven one thing, American society in particular, which is we're incredibly slow to adapt to anything in technology in terms of regulation, in terms of best practices, in terms of any of those things. Like the internet is still barely regulated ⁓ by comparison to how it should be most likely. ⁓

I mean, that's the thing. You can't regulate something that can't be regulated. The whole point of the internet is it's geo distributed. So, you know, we've seen this in the EU and I laugh about it all the time. Like GDPR exists and they have all of their laws and rules, some of which I think the intent is correct. The execution is wrong. We have people that come, I mean, we got somebody that tried on us and they were like, well, you dropped a cookie on my machine. That's in violation of GDPR. I'm like, or a US company that doesn't operate in EU, go fuck yourself. And you don't get to apply your rules here. We even have it within the United States, right? CCPA applies in California, but doesn't apply in the rest of the states. So at what point, how do you govern something that can't be governed? And we've seen this play out with piracy. We've seen this play out with a lot of things like, You can try, you're wasting your time. It's an inevitable game of cat and mouse and you're better off trying to figure out like, how can we evolve with it and not against it? Metallica tried this shit back when Napster came out, right? ⁓ The idea of online distribution. ⁓ And like ⁓ to their credit, somebody was gonna do it so they might as well do it. But in the end, online distribution won. ⁓ And it wasn't arresting grandmas and people for downloading stuff on Napster and LimeWire. It was Apple finally identifying that like, listen, guys, this is a wave that's going. So we can either keep trying to hold back the water and continue to get bowled over by it, or we can figure out how to ride this wave because people don't want to buy your bullshit CD that has two good songs on it. They just want to buy the one or they and now we just stream everything. We don't even buy any music.

So we started this conversation by asking the question, does AI make us dumber? ⁓ And I think we've ⁓ rounded third base maybe in heading home, which is like this idea of ⁓ maybe it could, it will make some people dumber. ⁓

⁓ Yeah, I think I started as like, yes, right? ⁓ think it I think topically, anything you use it for, it inevitably makes you dumber for that thing. Right? If if I use AI to help me develop things, or just to develop an application, it's going to inherently make me dumber about whatever that thing is. And maybe that's okay. I gave the example last time ⁓ of using AI to write Go CLI apps. Because I don't give a shit about writing Go, and I don't give a shit about the CLI app necessarily. It's a very simple application that just needs to do a very simple thing and be a wrapper around an API and work well. So don't care. I don't care to become educated on all of the nuances and details of that language. I can read it well enough. I can spot check it. I'm good with that. When it comes to other aspects of what I do, right? When it comes to aspects of things like Omarchi and ⁓ how Linux systems work and ⁓ how our business works ⁓ and advertising and technology in general, those are things that like, don't delegate those to the computer yet. ⁓ It's important to me that I have a deep understanding of how those work because that's how I apply ⁓ that knowledge to ⁓ whatever it is that I'm working on. But some aspects I'll happily delegate to the computer, and I just don't care. And I think that ⁓ that becomes the maybe side, right? So I start with, ⁓ it does make you dumber, but maybe that's OK. Because now I don't have to, instead of having to invest time, energy, and materials into learning Go to create a CLI app, I've just moved past that to something completely different.

And I think that's.

And that's the problem I'm actually trying to solve, because that was really just in the way. ⁓ I needed a ⁓ CLI that interfaces with an application that we built to be able to enable some automations that I want to be able to do. It was never about the CLI app itself. So yes, I am dumber as it relates to writing something in Go. But that's OK, because I'm smarter about the topic that I actually cared about, because I was able to take all the time I would have invested over here and invested over here.

So, I mean, you're describing the difference between critical thinking and subject matter expertise, right? So ⁓ critical thinking is figuring out how to accomplish something. Subject matter expertise is using the exact tools that you need to solve that problem. And I think where my head is, and where AI fits in this equation is the critical thinking part, least today, cannot be replaced by AI. It's very difficult to replace that. ⁓ will hallucinate after step three of a 10-step process. It just does. ⁓ so knowing what the 10 step process is, or at least ⁓ vaguely how to guide it towards that 10 step process is part of the critical thinking loop. What do I need to do? What are the likely steps I need to take? I can use my favorite brainstorm buddy in the form of AI to help me round out what those steps are. That's fine. But I still had to know that there had to be certain steps and know whether those steps were in roughly the right order or were missing altogether or whatever. That's the critical thinking part. And then the actual execution of it, I'll happily delegate that to AI. You know, and have, right? I've had AI help me create project plans for go-to-market launches and create individual base camp cards with certain deliverable sets. And then one at a time be like, all right, bro, we're going to go and create a whole bunch of deliverables for this go-to-market plan. I don't want to know how to design that. I don't even need to write any of this stuff. You have the brand guides. You have the tone of voice. You have all the things that we worked on as part of the critical thinking loop of this that I care about. And now I needed an execution buddy. And that's That's name, your name's Claude today. Sometimes your name is GPT. Sometimes your name is somebody else. ⁓ but that's, that's the role you're going to play.

When what you just described is all the stuff that like, it's patterned, like very patterned and pattern-matchable, right? We talk about that all the time, that like, current generation of LLMs are basically just really fucking good autocomplete. And if you have something to autocomplete to, it does a fantastic job at that. When you don't have something to autocomplete to, ⁓ It has no idea what to do with that. It hallucinates and just does all kinds of wild stuff. so we think about something like what you just described. All of the parts where humans provide value, all the parts ⁓ that really require brain power, are all the parts ⁓ that you talked about doing yourself. It wasn't until you get to a certain point where everything else is just executional, it's busy work that needs to be done, and it needs to be done the same way. ⁓ We have projects, projects have templates, they all run the same way, they all have the same rules, right? You produce something, you review it internally, you have a client review, right? That's formulaic, that's pattern-matchable. We don't need ⁓ to invest human cycles in doing that anymore because we have a pattern we can just apply over and over and over over and over once we've gotten it to a certain point. And ⁓ the same with creating those tasks, right? We set those tasks up in the same systems the same way every damn time. And ⁓ plenty of systems exist, right? Basecamp is one that we use. Basecamp has the ability to create template projects, which ⁓ we have. These have all been examples of how we're trying to do this already. But now we just have a smarter templating system where it ⁓ can recognize that this proposal has two rounds of ⁓ revisions, or this has two different creatives, or this has ⁓ ⁓ extra pages or I don't know what the hell it is. Anything that, ⁓ any nuance and can apply those variables and create project plans and things that ⁓ match that without using human capital to basically just do these repetitive monotonous tasks ⁓ that everybody would just rather not do anyways.

Yeah, so ⁓ question to round it out. Does AI make us dumber? ⁓ or ⁓ the answer is maybe. Plausible.

Maybe. ⁓

I don't know, is it dumber ⁓ to work smarter in some cases? Is it dumber to eliminate busy work? No.

Well, it depends on your definition of dumb is really where it comes down to. If dumber means I don't know how to do the executional pieces because I'm delegating that and I would anyway, right? I'm not, I'm not doing a lot of that execution myself anyway. I'm going to delegate it to a subject matter expert, a designer, a writer, a developer, whatever. I just so happen to now have the ability to do most of that without having another human involved. I'm just delegating differently. Right? So if dumber means ⁓ I'm losing the ability to critically think. I do think there's a major risk of that with people that have an overused pattern of AI, especially those that are content that are just now entering those critical stages of the human learning cycle. So junior high kids. mean, I worry about my kids like they're, they're, they're good with computers, but I, I worry about them, you know, growing up in an age of AI without having those foundational critical thinking skills and then not knowing what to do with them. You see it now with like, with higher education and colleges all freaking out that kids literally do not know how to write. And we've, we, it started with the social media, with, with the social media generation, right? Kids don't know how to write in ⁓ like complete sentence patterns. They're used to text form sentences or short form sentence.

Yeah, but neither do most fucking CEOs and they went through normal. Like have you ever seen like each CEO emails? They're like three lines long, all lowercase, no punctuation, right? Like who cares?

Sure.

Yeah, that's fine. But they already they they've proven that they already have the foundational elements done and they're just choosing to be lazy. That's different. Right? These kids are actually turning in papers that sound like they're writing to their social media friends. Now pre AI in thinking that it's good and fine.

Maybe, or...

Sure. ⁓ I think the tooling that we have in education is fundamentally flawed. I don't think writing papers is a good measure of success. And I think that's something that should be evaluated by, you know, formal education systems. I don't have the solution to it, but I do know that like, if I think back, like I'm a very critical thinker, right? Like we had some issue with our AC here and I was telling my wife yesterday about kind of how it went down and didn't deal with the techs. And I had already diagnosed a good portion of the problem. ⁓ Fix it but kind of done enough of the diagnosis that like ⁓ I was in in the ballpark of what the actual problem was before they showed up And it's a topic that I know nothing about. I don't do the HVAC work. I only know about it through whatever. But that pattern is the same critical thinking pattern that I apply to pretty much everything. And I didn't learn those by writing papers. I learned those in science classes and a lot of the IT work that I did. It's a lot of the like, formulate a hypothesis, test the hypothesis, find the result, and do it again. So I wonder if there's not some elements of those aspects that should be shined on more of like, you know, what are things that are experimental ⁓ and what are things that we can encourage people to do that like still hit the same muscles and still work the same way, but aren't as subject to gamification or ⁓ things that just don't matter. I don't know.

It's a hard subject, I'm glad it's not my job to figure it out. But we will continue to pontificate and have opinions. ⁓ For those that are trying. ⁓

is true.

Yeah, we'll just tell other people. We'll just tell other people they're doing their job poorly and they need to figure it out. The education system needs to figure this the fuck out. What's the solution? I don't know. You just figure it out. Just do it. That's what clients tell me, right? Just. We'll just we'll just do it. We'll just make it.

Email us if you want our opinion. ⁓ Till next time.

Yeah. ⁓