Focused - 262: The Strategic Deployment of Friction

Episode Date: August 11, 2026

Welcome to Focus at Productivity Podcast about more than just cranking widgets. I'm David Sparks and joined by Mr. Mike Tritz. Hey, David. How's it going? I'm great. It's hot, though. I'm in Southern California and we're now in the hot dry spell, but it's a little wetter than normal.

We have a little humidity this year. Oh, you, Mike. I always hear about how cold it is in Wisconsin. I don't know. What's it like there in July, August? It's all over the place. We just had a heat wave where it was like 100 degrees for an entire week. week and yesterday the high was 72 so yeah wow i don't know you never know what you're going to get in the

midwest yeah well either way i'll tell you where it's hot it's hot in the ai space everybody can't stop talking about it what they're doing with it and uh we wanted to to discuss it in relation to focus and today's episode is about that topic i feel like uh in preparing for this episode So it's been a great opportunity for me to kind of think through this stuff. And I think we all need kind of a personal AI philosophy. And our goal today is to help you structure that and have some good stuff to base that on. Yeah. And the inspiration for this episode, which I think is going to make it a little bit different

than the standard, let's just talk about the AI moment thing, is I recently listened to how I write podcast. That's David Perel's podcast. and he interviewed, I think, 12 different experts about how they think about AI. And so there's a lot of differing opinions, and they're kind of juxtaposed next to each other. So you get people talking about it at different points. And this is the How I Write Podcast, so it's the creative process. But as I was listening to that, I was thinking a lot of the things that they're saying and a lot of the takes that are on that podcast apply to the general focused productivity spaces as well.

So I think there's a lot we can learn from those perspectives about how we're applying AI or not applying AI through the lens of how do I do more of what really matters. That's the whole focused angle. But I wanted to use that as a jumping off point because these are some really smart people who are the best of the best at what they do. And they have very different views about where AI fits. And that just hits on the point that you made. I think you've got to develop your own personal philosophy about this stuff.

It was interesting because it is like getting racked around the inside of a car going around tight turns because you'll go from Sam Altman to Ezra Klein and they have very different opinions on what you should be doing with AI. Yep. And none of them are the canonical right way to do it. You have to figure out where to draw the lines for yourself. And I'm thinking that's going to be the theme that's going to emerge here. and for each of the sections that we've got in the outline,

as it applies to different parts of the process that they were talking about on that podcast, I think that there's a lot of great discussion that'll come as we consider for ourselves. Where are we going to draw those lines? Where do we start, Mike? Let's start with AI for Research. So this was the first part of the episode. This was Ezra Klein. And the big idea here is the, he talks about the act of erosion.

He has a line in there that AI isn't bad. It's specifically good at the parts of the work that he most needs to do himself. So he's recognizing that AI is good at this research stuff, but also that's the valuable part of the process for him. And he's trying to figure out, you know, what do I actually delegate? What do I hold on to myself? And this was basically a different version of this question is something that I posed to James Clear at the Craft and Commerce conference.

And it's like, where do you think about AI in the writing process? and any parts of it that are sacred to you. So Ezra is basically saying that the actual research and putting together the RAND report is the example he used, that is sacred to him because of the friction that comes along with that being the mechanism for leveling up the skill. I think that's really the overarching theme here is, as we're talking about when to use AI,

how do we stay sharp? How do we not become just the dull knife in the drawer? as we let it just eat up all this stuff as it becomes better at different things. And I think that's a pretty interesting question. You know, where do you draw the line? What friction is productive and what isn't when it comes to the work that you do? Yeah. I, you know, the big point I got out of his portion is that it's only when he lives with something for a long time that he truly starts to understand it. You know, the process of reading several books on a topic means that that you've got the topic running in your subconscious, you know, for days or weeks.

And that's where the insight comes from, from living with it. And where if you have a magic box that just gives you the research and you can just read a nice summary and be like, you know it. You don't know it as well, but you also don't have the benefit of the sub process in your brain going for weeks or months. And as a result, you don't get as much insight. And I thought that was very true. I mean, I think that some of my best thoughts come from stewing something over for a long time. And I hadn't really thought about it in respect to AI. It doesn't do it.

For me, personally, AI does some research, but usually it's introductory research just to get me to the right spot. The big argument of my productivity field guide is virtue-based living, you know, that you live based on a virtue system. And I had several people writing me from different religious traditions saying, you know, we have something kind of like, that in Judaism or, you know, Islam or whatever. And I thought, that's interesting. So I had AI prepare me a reading list, but I did the reading, you know, and I don't think I would have got to that reading list as quickly, but I was the one who went and kind of explored it. And I do think there's something to that, you know. Yeah, and that's the fascinating thing to me about all of this is AI definitely reduces the friction, but sometimes reducing the friction is good, sometimes reducing

the friction is bad. I like the term intentional friction. And I feel like that is the focused approach is adding friction back into the process at specific points in service of a better outcome, a better end product. And that is tricky. There's no simple formula for that. But that question is insanely valuable. And I didn't really consider the erosion idea that Ezra was talking about in this section where he made a comment about what changes you isn't the wave, it's the erosion. So when you use AI as the easy button, there are tradeoffs with that. And it's okay to do it in specific places intentionally when you need it to help you get started with the research as you were talking about, David. But if you're constantly just using it and taking what it gives you as

this is, you know, I'm done with that part of the process, I feel like the real detriment isn't the fact that you did it once. It isn't the fact that you did it a couple of times. It's the fact that you now have a new norm and you're taking the easy path, the path of least resistance, every single time. And if you do that consistently enough, then your focused muscle atrophies, your ability to think deeply and critically about things atrophies. and you should be considerate of that. But I don't think, you know, if you use it at all, that it's necessarily a bad thing either. I think actually it can be a net positive if you use it for certain parts and then recognize,

okay, this is where I have to intervene and use the gray matter between my ears. I think that it can actually be a net, that can be like a focus booster in a way because you're being focused on the thing that you provide maybe the greatest benefit or the greatest value to the process and you don't have to worry about some of the minutia. And that's grossly oversimplifying it. But figuring out where do we apply the intention and where is it not all that important? That's a really big question. And the two opposing forces in this whole subgenre of AI is the,

the ability of AI to completely remove friction and make things easy compared to what we are, I guess I would call it the strategic deployment of friction, realizing there are things you can do or actually having friction is good. I was talking in a workshop just yesterday, and one of the participants was talking about how he likes the analog last mile for his tasks, because if he doesn't finish it, he has to write it down the next day. you know, and that friction actually helps him make better decisions. That's not really the same concept, but I do think in this world right now,

we have a revolutionary opportunity to remove friction from a lot of process. And I think we should be thinking about where that does not make sense, you know, even though the robot, the AI, can, you know, give you this frictionless life. there are parts of it that you need to retain the friction and embrace it. And being smart about that can make a huge difference. Yeah, that's the trick, though, right? Is identifying where to dig deeper and where, you know, this is good enough.

I feel like that's the difference that the next person on the podcast, Tyler Cohen, talked about. And these are kind of juxtapose at the beginning, because they both do similar sort of work, but Ezra Klein, you know, he wants to do the research, and Tyler Cohen is all in with AI as the new secondary literature, how he describes it. So they're both doing the hard work,

they're reading difficult things, but they're kind of reaching opposite conclusions on whether AI is valuable in that part of the process. Now, I think, though, the key is in the detail, here of how Tyler uses it because he'll use AI to interrogate what he reads. And he says in the podcast that that actually makes him a better, more active reader. And he makes the point that people who are using AI and they're having a bad experience with it at this point of the process, they are bad users who are asking bad questions

that are too general. And I have definitely noticed that myself. If I get more specific about what I am actually using AI for. than the results that I get are better. And I'm interested in this. I don't actually do this sort of thing, but I've heard of people who have done this where they'll read a section

and then they'll just ask AI to grill them on it. And I feel like if we're really trying to understand a concept, that's actually a pretty interesting approach. You know, I send a newsletter out every Saturday and that's my best writing, I think. And I actually, when I finish the draft, I have the AI ask me reflective questions on it. And that has over time changed my writing.

And the prompt is not rewrite it for me or fix it. It's just like asking me reflective questions about things I missed. And I feel like there were certain tendencies in my writing before I had that coach that were not good. And I feel like over time, I've been doing this now over a year, it has improved my writing. And one of the things, I think it was Tyler who said this. I wasn't clear always who was talking. Yeah, it's a bunch of cuts.

It's really hard to follow sometimes. Yeah, but one thing she said that really resonate with me is the AI is like having an editor that's always on call. And a lot of times I'm writing in the middle of the night or early in the morning. And frankly, I just don't have that editorial relationship with someone that I feel like I can always just send them everything and have them check it. And I do find that it really is helpful. And like one of my favorite prompts in AI is like, I'm about to send this to the internet, look for any showstoppers. And my A understands that doesn't mean I want you to rewrite it for me or go and tell me that this is stylistically wrong. And because sometimes my stuff is grammatically wrong and that's a choice.

But I just want to know if there's anything that's going to make me look stupid, you know, and occasionally find something, you know. So that really lands with me. As much as I feel like, you know, we do need to be intentional about friction. I do think it improves my writing. I get these emails on occasion for people saying, I think you're using A.A.A. to write because I was looking in your posts are longer than they used to be. I'm like, that's because I have more time to write now. It's not because I'm using the AI to write.

I'm actually trying, if anything, it's giving me more time to be kind of in that mode. I don't have to be chasing my tail, putting blog posts into WordPress and all this other nonsense. Yeah, and that's the difficult thing is that question really isn't nuanced enough in order to act in order to really answer it. Do you use AI for writing? Well, if you use it for spell checking grammar, technically you do. But really what they're saying is, did you use it to generate words that I am reading? Yeah. Right.

And I think that even that is, I think that's nuanced. Because if you have a general idea of what you want to say, but you are stuck staring at that blinking cursor, that is a point of friction just to get started. Now, the writing, there's friction in that whole process. And I do think that actually formulating your thoughts as you write them out, there's value in that. You know, thoughts to entangle themselves, your lips and pencil tips. And I would add also clicky keyboards. like that's the general principle there. But you can't revise something that you have,

haven't written. You need something to respond to. So I actually think if it can help eliminate some of that initial friction and just get you going, prime the pump, go ahead and use it. But the thing that I really liked about that phrase, you said having an editor that's always on call. The always on call part is the one that I completely resonate with. I remember there was a moment when I was,

working with Asian efficiency, and that was the product guy. So I would go down to Austin every quarter, and we'd talk about what we're going to build, and that was always really, really energizing for me. I am a very big introvert frequently at home. I'm like, Rachel, I got to get away from here. There's too many people. But in that setting, where we are collaborating and we're bouncing ideas around, I could do that all day. And so we're there. We're talking about the product that we're going to build, which is like my wheelhouse. So I'm probably more invested in it than other people there.

But that's really exciting to me. We talk about it through dinner. And then I'm staying with one of the guys on the team. So we're at his apartment later in the evening. And we're continuing to talk about it. At one point, he's just like, Mike, I'm done. I can't talk about this anymore. I'm so spent.

I need a break. And I was like, wait a minute. You're the extrovert. I'm the introvert, like, what's going on here? Right. And so I realized that there's a different, different type of energy that comes from that, that creative process for me. And that collaboration, and specifically, there is something magical about that for me.

And some of my best ideas have come from just, I'll book a room at the local library. They've got glass boards in there. And I'll go in there for like three hours. And I always put on the reservation thing brainstorming session. Yeah. And I'll get on the whiteboard and I'll just. start doodling and then I'll have cloud open on my my MacBook next to me and what do you think about this? You know, I'll take a picture of what I got so far. Like what holes do you see? What ideas do you

have? And it'll say, well, you could do this, this, that and the other thing. Oh, I don't like three of those, but that one right there, that's a really cool idea. And it sends me off in a million different directions. So in some sense, I think that AI is kind of the ultimate creative collaborator because it's never going to tap out. As long as you want to, I ideate and bounce things back and forth, it'll be there for you. Yeah, and yet, right, it's just a word algorithm. You know, it's not a person. It's true.

And, like, I'm always aware of that. Like, one of the, people asked me, because I made that course, they're like, why you keep calling it a robot? Why don't you give it like Gs or, you know, some fancy name? And I want to just be very clear in my head that this is not a person, you know? And I don't know. I understand what you're saying, but I reach a limit with it. in terms of idea collaboration.

Look, I do talk to it about ideas and sometimes it gives me feedback, but I don't think of it. It doesn't rise to the level of creative collaborator for me. It's more of a tool. It's like this is the thing I wanna do, this is the direction I'm going to use any holes in it. That's like my typical, like when I'm at that phase,

my use for it. And neither one of us are wrong. Right, right. I think one of the things that has gotten me to the point where I'm okay embracing it as a creative collaborator has just been the last couple of years with what I've done with YouTube. Because I am not naturally a YouTuber. I would say I'm a reluctant YouTuber. I got into YouTube because Matt Reglin was coaching me at the time. I had this idea for Life HQ and he's like, dude, you have to start a YouTube channel. Like, oh, okay, fine. And then

he pushed me further. And he's like, you need to create a video every week. And I'm like, I don't know how to do that. I don't have that in me. And it's true. If I had just left it to myself, you know, I wouldn't, I would have ran out of ideas. I would have been exhausted from that process.

So over time, in an effort to level up my YouTube game, I have started to study the different parts of the YouTube process. One in particular that I feel I am just not wired for is writing YouTube hooks. I am not good at writing hooks. Fortunately, there are tools out there now. So when I have the idea for a video, I can hop on to vid IQ and say, you know, this is the angle I want to take. What do you think I should consider in terms of like

titles ideas, thumbnail ideas, hook ideas? And it'll come back because it has access to all of the videos on YouTube. It looks at which ones are outliers, the ones that are performing well above all the other ones for that particular channel. And it'll say, you know, this one performed pretty well, here's what they did. It's not a science. It's an art. You kind of look at what works and try to tease out what is it about that that actually worked. But if you look at some of the patterns that you see on YouTube, there are different ways that people do certain things. And some of it I'm like, no, I'm not going to be that clickbait. I don't want to, you know, you'll never believe. You know, I don't want to do that. But I see the structure they used over here. And then that'll open up a loop.

oh, I see how I could apply that to PKM and Obsidian, and I'll take it in this direction. And when it comes to, like, writing the hooks, just because those first 30 seconds are so critical for YouTube, you got to really say the right things in those first 30 seconds. My tendency is to summarize everything, and my hook is about two minutes long. And by that point, you know, people have left. So I will ask it to give me, write some hooks for me so I can look at them. And I never use exactly what it gives me, but it, It knows the concept behind the video.

It knows the outline, the structure, and it does a much better job of writing those hooks than I would if I made them myself, no matter how much time I put into them. I would never get to the point where, yeah, this is a good, good YouTube hook. And I can take what it has, and then I can make it my own, because there's always a few things in there, like, I would never actually say that. So I'm not going to, I'm going to edit that. The real key for me is I want every word that I show up and say to a camera to be authentic, to be something that I would have said myself.

Yeah, I think one of the keys, you know, I have a separate podcast intentionally, where we talk about stuff like this every two weeks. And one of the things that I continue to bounce on is this idea of context. And I feel like for the creative process, it needs to fully understand your context of what you're doing and your voice.

And I just think is really hard, at least at this stage of technology for it to have sufficient context to really do that with any effectiveness. So sometimes it may randomly stray onto something that's useful to me, but I really, I really try to like keep that to myself largely. But yeah, I get it. Like at YouTube something you don't have a lot of expertise in and it is good at getting you to a certain level. But as you advance in your YouTube production, you'll get to a point where you know, know that. I know what my hook is because I know what my voice is and it doesn't know me as well as I know me. And I think I could see you kind of graduating beyond that at a point.

One of the points they made during this, I believe it was during this podcast or I read somewhere else. I've been thinking about AI a lot lately. The person who's talking about grammarly and how grammarly does a great job of bringing you know, subpar writing up to a certain level. but at a certain level, it actually pulls down excellent writing because, you know, you get to a point where maybe it's a little odd the way you're putting this sentence, but it uniquely makes the point where grammar really will immediately try to like, you know, take the sharp edges off and take all the life out of it. Yeah, so that was, I think Stephen, no, yeah, Stephen Pinker. He was talking about how if you fed large language model all of these faces from a yearbook,

the composite face that it would take that as it averages all that stuff out. If you morphed a few hundred these yearbook photos, you would get someone who is better looking than any individual, but the result is generic. Yeah. Right. And then Gramerly basically makes the bad writing better, but it also makes the great writing not as good. And so that was sort of Robert McFarlane's thing,

because he was sharing some of his prose. He had a couple examples of verbalist paragraphs about a Canadian river that he fed into Gramerly. and Gramerly absolutely hated it. But that's his style. People like that. People like the personality. Right.

So those are two things that Gramerly is doing, but it's really just one function. It's trying to get it more like a specific thing, following a specific pattern. So the larger point there is if AI raises the floor and it lowers the ceiling, you got to know how and when to make sense to actually use it. This episode of Focus is brought to you by Scribe. AI-powered workflow documentation.

Get your first month for free. Go to Scribe. Dot how-slash-focused. Scaling isn't a headcount problem. It's not a budget problem either. It's a documentation problem. Critical knowledge stays locked in people's heads because writing it down by hand

is too much friction to ever get done. The result is inconsistent workflows, tools nobody fully adopts, and when someone walks out the door, their knowledge walks out with them. It adds up fast, wasted hours, underused software, and no way to turn your best practices

into everyone's practices. Which is exactly the problem today's sponsor, Scribe, was built to fix. Scribe is a workflow AI platform that's trusted by 94% of the Fortune 500. It captures your workflows in the background as you work and automatically turns them into clear documentation.

So your term can say goodbye to writing everything up by hand, stitching together screenshots, and recreating the same onboarding docs every time someone new joins the team. As I run Max Barkie, there's actually a bunch of administrative work that I can't really trust to an AI and I need a human to help me. The problem is I'm too much of a perfectionist and I want things done just right, but I'm not willing to put the time in to train somebody. That's where Scribe comes in really handy. It will capture that process and document it for me. So the next time I need to provision a customer, or delete an account, the Scribe process will give my assistant exactly what he needs,

and that's exactly the kind of process Scribe can capture as you work. Getting started is easy. Scribe captures your workflows as your teamworks, turning them into step-by-step guides, screenshots and all automatically. What would have taken hours of writing, recording, and cleanup is ready to share the moment you finish. Scribe automatically redacts sensitive information, names, account, numbers, emails, from every screenshot. As an admin, you can enforce this across your entire team. Feel confident nothing slips through the cracks. Anyone following this process can launch real-time, on-screen guidance that shows them exactly where to click step by step inside the actual tool.

It's there the moment the Scribe is created and ensures everyone does critical processes correctly. Scribe will even suggest improvements to the underlying workflow, not just documented. Once you can see how a process is actually being done, you can identify for, where steps are redundant, along with where people are getting stuck and where things could be automated or simplified. To see what Scribe could actually look like for your organization, head to Scribe.how slash focused and get a free first month on select plans. That's S-C-R-I-B-E-D-How-S-Focused. And our thanks to Scribe for their support of the Focus Podcast and Olive Relay. So one of the more interesting speakers on the show was, I guess, Jimmy's so much.

I've never heard of this person before, Mike. I guess I'm out of touch. But much more enthusiastic about AI in writing. Yeah, there were definitely some people on this podcast that I was not familiar with their work either, but they were all very smart. Jimmy mentioned that he thinks of AI as the most unbelievable tool I've ever seen to improve every single part of the work that I do. And he had a very specific example here, which, again, I completely

resonated with, he mentioned that his daughter writes freely because there's no self-consciousness, self-consciousness between the impulse and the act. And I have been at that place. I kind of described this a little bit earlier with the blank page syndrome. You've got this idea for something you want to make. You don't really know how to get started. And so you sit there and you stare at that blinking cursor and you feel worse and worse and worse about your ability to write and create until eventually you get to the point as I did where you think to yourself, I guess I'm just not creative. So he mentions that AI is the anxiety antidote. I love that. If you are stuck, use AI to get unstuck. Don't use it always as the path of least resistance, but using it to get

the, I have this framework, to creativity flywheel, get that flywheel spinning again. I think that's a great idea. And the key is to recognize that he has this phrase in there, AI is middle to middle, it's not end to end. I think that's really important. You can't use it to do the entire thing. But if you can intervene at specific points and apply it intentionally, then the end product can be better. And it doesn't have to lose your voice. It doesn't have to lose your taste. It doesn't have to lose the personality. It doesn't have to be something that is inauthentic. It can still be completely yours, I would argue, even if you use AI at specific points. in the middle of that process. But you have to have ownership over the whole thing from beginning to end.

You know, I think it's because of just my life experience. I honestly have never had writer's block. I've never really experienced that anxiety involved with writing. And I think part of it is because I made my living writing from a very young age. You know, when you're a young lawyer, all you do all day is write briefs. And, you know, you got to get them out the door or you don't have a job anymore. So I got very good at it. And so I don't really feel it that way. But one of the things he said that resonated with me was just using whisper flow to get the lousy first draft in. But I've been talking about that for decades.

I always have felt that boy's dictation is an easy entry point to get started writing if you're having trouble. But overall, I actually agreed with almost everything he said. I think this stuff can be an excellent tool to help you, you know, improve the process. so long as you're careful, you know. Yeah, and the example of whisperflow, I love that because I've heard you talk for years about Dragan dictate before that. And I feel like you've always, the way you've described it, where I process better, I think better when I'm talking out loud.

That's a very astute observation. And people should recognize whether they think better, you know, verbally, verbally or they're cranking out words. or they're sketching things, but that is eliminating the creative friction. And I don't think when you started using Dragon Dictate back in the day, that you would have said, I am using AI to write.

And that's really what Jimmy was saying at this point is, I'm using this tool, which I would argue, yeah, Whisper Flow especially is AI because it's not just your verbal, like whatever you say. It's trying to clean it up a little bit, make it a little bit more coherent. So there is some AI, being applied to it, but even the translation of the spoken audio waves into text, that is AI.

We don't think of it that way, though, because we think of it as this generative AI where it's just going to fill in the blank for the next token, and it's not that simple. There's a lot of nuance to this. There's a lot of depth to us. There's a lot of detail and color. And I feel like you got to really spend some time dialing this in. But if you, you do, it can give you, I mean, not just creative, but I think it can give you creative superpowers, but really take that broader productivity. And like you want to focus more on your, only the things that you can do, the deep focus work, do more of what really matters, right? Stop cranking the widgets. If you could use AI to crank the widgets, you could do the life

stuff. And that's, that's really the gist of the, the whole argument, I think, here. from this whole episode is how do we stay sharp? You know, you mentioned the strategic deployment of friction, right? Yeah. Where do we do that in order to develop our skill and maximize our A.T. To borrow the term from the productivity field guide, you know, how do we become the best version of ourselves? AI is a great tool for that, just like any other tool before it, but it is a tool.

Yeah, I mean, the thing through my dictation workflow, which kind of predates large language models historically is I would before I had any dragon dictate because as a baby lawyer, they hand you a little tape machine. You put physical tapes in, you press the record button, and I'd walk around my office and talk through a brief. And then we had people that would type it all up for me and I'd read it and then start as the lousy first draft. And over time now, that has improved where we get immediate response. It doesn't have to be typed by human. And now we've got AI and it adds a new layer to it. So what happens now is I'll talk through an idea.

And AI is instructed by me to ask reflective questions on it. Again, that reflective question thing is a constant in my life with AI. Because I feel like to me that's my comfort zone where it's like, yeah, use the word machine to see if it pulls something new. out of me, which I wouldn't have thought of. And that's really useful, which I hadn't done in the past. Another layer for it could me to say, okay, take the 2,000 words. I just dictated it and boil it down to 800 that are efficient. And that's where I don't do that, you know, but I could see you using that. And you could argue those are your words because you're the one who wrote the 2,000 words.

and then the AI was the editor that pulled it down to something more tight. But I like, to me, the little edits are really important. But the reflective questions are what I do. But, you know, I don't think we're making this episode to say, this is how you do it, but just to open the questions. Yeah, you got to open the questions. And the big question, I think, with all of this stuff, is what piece of ownership do you take in the process?

Yeah. There's a quote from that Jimmy Sonny section, it's the sharpest knife an engineer has ever been given, but it's still a knife. You still have to cook the meal. Right. And I remember the first episode, I think, of intentional AI was something like anyone can cook from the Ratatouille movie.

Yeah. Right. I think that's the cool thing about the tools that are available is it eliminates a lot of the barriers to entry for a lot of things that traditionally there was there was quite a moat there. And you can look at that from one perspective and you can say, oh, this is a terrible thing. You know, I was just listening to a podcast the other day about how there's all of these vibe-coded apps in the app store. I know you've got one.

I've got one that I'm working on. And they're like, this is a bad thing. You know, all these apps that people don't know how to develop, they're releasing these things. I don't think that's a bad thing at all. I think it's a, I still think that if you compare my vibe coded app to somebody who is a legit developer who's got a good idea and can apply their expertise to handcrafting their app theirs is better. And if they are solving a real problem that the market, in this case the app store is having,

then they're going to be more successful than I am. That's okay. But I've always wanted to make. this thing and I haven't been able to make this thing and now I can make this thing. And it doesn't have to be a huge financial success. I'm not trying to make my living making apps on the app store. But if I have an idea for the specific app I want to make that's different in a couple of different ways than anything else out there, I don't have to bombard my favorite developer with feature requests anymore. I say, hey, add this thing, add this thing, add this thing. I can build the exact tool that I need so

that I can cook. So there is a dark side to this. And kind of let's talk about Providence and some of the stories out of this episode that I thought were really well received. Yeah, this is kind of continuing on the ownership idea. But Providence is a word that means the place of ownership. And the story here is from Jan Martel about the, they were, he was at a wedding and at all the tables, they had these handwritten notes from the bride and groom, and he went up to the groom

in the middle of the ceremony. Well, this is a little deeper than that. The note was about why your presence at the wedding was appreciated and why you're important in their life. Yeah. Yeah. Yeah. So then finds out basically we had AI write those.

And he said it felt like a violation and a betrayal. Yeah. And I can understand. understand that perspective. I haven't had that experience. Sounds like maybe you have. Yeah, I have. I went to an anniversary where a toast was given and it touched me. You know, the person who gave the toast, I felt like, my goodness, I didn't know he had it in him, you know. And so I pull him aside later and just said, look, I just want to tell you, I really enjoyed your toast. And he's like, oh, I had a, I read it. And I just, you know, you just feel there is a sense of violation, like, because I open myself emotionally

to those words thinking they were from his heart and they weren't, you know. And that I think is the soft underbelly of all this stuff. I mean, you know, there's a betrayal to your audience if you're, if you're doing it wrong. Yeah. And even when I read that story, like I don't have that personal experience to anchor it off of. But what really stood out to me from that was the nonchalant response. Yeah. From the grooms, like, oh yeah, I used AI for that.

Yeah. almost like a point of pride. Yeah. I made it easy. I took the easy way out. And see, I wouldn't have approached it that way. I still don't think I would do what they didn't. I think there is something about the words that are coming straight from the heart, whether they are not as clean and organized, they are more valuable in that scenario. So I don't think still I would use AI in that scenario. But I do think there's a version of that where using AI is actually a net positive,

where you're doing that for hundreds of people, right? You have this picture in your head at the beginning of the thing you want to deliver. I would love to have a handwritten note for all of these people. I don't think I can do a handwritten note for all of those people. I need a little bit of help

for each one of these. Don't let it write the whole thing. I don't think I would ever do that either. But I think there's a scenario where if the alternative is, there's no handwritten notes. There's a version of, I used AI a little bit in this process to create these handwritten notes and the words that

I am sharing are authentic. And I don't tell you that, oh, yeah, I just outsource this to the AI because it's not important to me that that actually is a net, net positive. But it's a very, it's a very gray area. See, that's a line for me. If I was going to give a, you know, a thoughtful note to someone, there will never be any, or I won't give one, you know. Yeah, I guess the version.

Having been on the receiving end of it, I do not. not want that so but again we all have lines on this stuff yeah so let me let me add a personal story here from my own wedding yeah um my brother was my best man and he was freaked out about giving the best man speech yeah so he talked to my dad and my dad said this is what you do gave him the script right you he's like uh so he stands up he's about to do his speech he's like rachel welcome to the family we're glad you glad that you're here but before you go too far with this. There's a few things you should know about Michael.

And he guess it's like big roll of banker paper. And he like rolls it out and it like keeps rolling and it keeps rolling. It rolls under, you know, all the, all the tables. And then starts like sharing some, some funny moments, right? He didn't come up with that. I didn't care. But it had a human source even though it wasn't your friend. It did have a human source.

Yeah. Yeah. That's true. But I, I don't know. Like he got some help. and I didn't ever think that, you know, he didn't actually say those words. It's kind of the same thing for me with, I studied Ali Abdel and his productivity, or his YouTube workflows. He's got a script writer. He's got a channel manager. He's got a thumbnail purse. He's got all these people

who help him with the process. He shows up and he reads the scripts that somebody else writes. But I also know that he has some specific slides that he reads out loud before he records a single video about how the metrics don't matter. And really the thing is about helping the person on the other end of this video who is investing their time to do this. And I think that comes through. No one looks at Ali Abdal's videos and says he's inauthentic. He's not, he doesn't mean what he says. He'd somebody else write those words. But for me, it's sort of the same thing.

If an AI model or a human is putting words in my mouth, I think that there's the warning sign, but it isn't necessarily the place to draw the line. Like, oh, well, that's inherently bad, at least for me. I don't know. Okay. So I think we're mixing, though, things. Like a YouTube video is very different from a wedding toast or a personal note of gratitude. I'm allergic to false sincerity, and I feel like, you know, when you get into the types of things like Jan was talking about, you know, there are lines that are clear to me, and that's just one of them.

But, but like I'm not faulting Ali Abd al-Ov or even having AI help write his scripts. He puts a lot of stuff out, and I'm sure he needs help with all of that. But I'm sure he's choosing what actually gets read and if there's something that comes. up that he doesn't agree with, he's not going to say it. Yep, absolutely. But, you know, it's like, like, this story to me is like a clear example of the line too far. But it's not for everybody.

Like you said, you know, some people proudly stated out. I was thinking both examples were men, I think often men are a little uncomfortable, showing emotional, you know, vulnerability. and that's like a great way. Like maybe he really meant that toast that he said, but he didn't want to say that and said, he said, well, hey, I wrote it, you know, and that allows him to kind of avoid the,

you know, the emotional depth that you would need to own up to it. But either way, it just felt really weird to me. And I felt this weird sense of violation because I was really moved by the words thinking they came from a human and they didn't. And that's the thing for me, I think, is by saying, oh, I had AI write that. Really what it is, it's an abdication, abdication. Abdication of responsibility and ownership.

I didn't write that. None of those things were mine. If you own it, I think it's, I don't know, I'm not going to say that specific, scenario is fine, but I think it's a lot more nuanced. I can see the beginning in the end of this particular scenario, and I don't like it, but I see a lot of things in the middle where like, oh, yeah, that would actually be, that would actually be useful. And I think that would be positive, like very, if there's a spectrum, right, and on one end is AI is not going to help me with my relationships at all, which is a different way, I think, of saying kind of what you're saying.

Correct me if I'm wrong. But on the other end is I'm going to actually be open and explicit about how it is helping me with my relationship. It's going to write the toast at the wedding, whatever. Like that's the spectrum. There's something way over near the beginning, which is I'm going to use AI to help give me triggers to apply the human touch. Yeah.

You know, like our friend Jesse J. Anderson has an app. It's called Wave. and it's literally built because we forget about the other people in our lives. And it gives you reminders like, hey, it's been a while since you talked to Jesse. You should reach out to him. You know, the fact that I have an automated reminder, AI telling me, it's been three weeks since you talk to your pal David, you should check in. You know, I feel like that's in the middle of that spectrum, but it's not, you know,

It's not telling me what to say. It's not writing the script. But even a little bit of, hey, go check in on that person. I feel like that's an example of the intentional way to use this stuff. And honestly, I have reminders in my system to remind me to talk to friends. Is that AI? I don't know. But I do think that, you know, there are some examples.

And I think you need to consider the audience. Like I was thinking as you're talking, if I had a problem with my wife and I wanted to talk to her about it and I had AI write the letter or script or whatever for me to talk to her, that would be too far for me. But if I said, I'm trying to address this problem with my wife, help me figure out the best way to communicate with her on it, I think that would be okay. So, you know, this stuff is all complicated. Yeah. And it's not just AI, I think, either. So what if your wife was really, really busy? Yeah.

And if you wanted to talk to her about some issue, you had to go through an assistant. Yeah. That would feel icky, right? Because there's this inherent social contract. I feel like that's at the heart of the wedding story is by saying, oh yeah, I had AI write this. You have broken the social contract. And what people really want in that scenario is connection. They want, but I think there's another version of that with even the creative work that you and I do.

And Sam Altman, actually, of all people, was on this podcast as well. And he talked about how the best selling novel of 2027, I mean, you can probably guess where Sam Altman falls in terms of where AI is useful, using it for everything, right? But he's talking about the best selling novel, 2027 is going to be written by a human because people want to know about the person who. wrote it. What people want from art is connection. What people want from relationships is connection. And to Jan Martel's point, he said he'd rather read a bad short story by a human than a good one by a computer. But I think if you know that going into it, if there's no surprises, no gotchas, the terms of the contract are clear. I think it's a little bit different because you don't have that scenario where it's like, oh yeah, I got tricked. And that's the heart of this stuff for

me is, you know, I'm trying to be open and honest about where I'm using AI in the process because I don't want anyone who watches any of my videos to ever feel like, oh yeah, you got me. That one wasn't really Mike, and I can tell. This episode of the Focus Podcast is brought to you by Squarespace. Save 10% off your first purchase of a website or domain using the code Focused. Just go to Squarespace.com slash focused. Squarespace is the all-in-one website platform designed to help you stand out and succeed online. Whether you're just starting out or scaling your business, Squarespace gives you

everything you need to claim your domain, showcase your offerings with a professional website, grow your brand, and get paid all in one place. I've been a Squarespace subscriber for years. I really like the service because they give you so many professional website features at such an affordable price. Putting yourself on the internet can be difficult, but with Squarespace, they make it super easy, super powerful, and super affordable. One of the things people want to do with a website is offer services and get paid. Squarespace makes that so easy. From consultations to events and experiences, you can showcase your offerings with a

customizable website designed to attract clients and grow your business. Think built-in appointment scheduling, email marketing tools, and more. Plus, keep everything cohesive with on-brand invoices and get paid easily with online payments. You don't need to jump to a bunch of things. to third-party services, Squarespace has got the whole thing there. And if you're running business, you need to track analytics since Squarespace does that too. You can keep track of the stats that matter with intuitive, built-in analytics tools. It makes it super easy for you to review website traffic, learn where to focus engagement,

and track revenue from bookings, invoices, or product sales. They've really got the whole enchilada there for you. Head over to Squarespace.com slash focus for a free trial. and when you're ready to launch, use the code focused to get 10% off your first purchase of a website or domain. That's Squarespace.com slash focused with offer code focus to get 10% off your first purchase and to show your support for the Focus Podcast. Our thanks to Squarespace for the support of the Focus Podcast and Olive Relay. One of the interesting things about this episode was, you know, I think a lot of people mix apples and oranges when they talk about AI.

and like one of the things is Ezra Klein was a little holier than now about it at some points. And I was thinking about him. And I was thinking, you know what Ezra doesn't need AI for? He doesn't need it to put his post on the New York Times or to get his podcast links together because he's got a team of people that does all that for him, whereas I don't. And I can't be holier than that. I need AI to do the donkey work for me. And nobody's talking about that.

I understand the show is focused on writing. But I do think when a lot of people can tell you how AI is, you know, useless or something you should avoid at all costs, a lot of times those people have humans doing essentially what AI does for those of us that can't afford humans. They have delegated parts of the work. And that's the thing that I want people to think. about from this episode. Yeah, there was a lot of focus on creativity and writing, but I think that's a that's a great example because that is maybe the sharpest example of you have to protect

the ownership, the authenticity, the originality. But whatever you do, if you are a knowledge worker, there's an element of that to the work that you do. And it doesn't mean that it's bad work or or AI slop, if you have delegated parts of that to either AI or other humans. It's perfectly okay to ask for help, but don't give up the control. Don't give up the ownership. Recognize what makes the difference in the work that you do. You've got that term moving the needle. We've talked about that in terms of journaling and making sure that we're following

through on the things that are important to us, I feel like there's a version of that that applies to the focused work that you do, whatever industry or organization you're a part of, whatever niche, there is, you have to find that balance between this is the value that I bring to this process and I'm going to maximize that. And the other stuff just isn't that important. So at the very heart of all of this AI stuff, it is a focus issue. I do think it's interesting, though, that everyone on this episode, as they're talking about it, talks about it at different stages of the work. You know, they've reached their conclusions and their opinions and they're sharing, you know, this is just takes from episodes that David Perel has done. So it's kind of cherry-picked

by him. But still, you've got Sony who's talking about drafting. You've got Cohen and Klein, who are talking about the input and the research. You got Maria Popova, Jan Martel. They're talking about the finished artifacts and the relationships to the reader, Pinker and McFarling talking about style. And it's, they're all describing different parts of the elephant. You know, there's that story about what is an elephant. Oh, it's, it's long. It's a, because one person's holding onto the tail. One person's holding onto the ear. And they're describing what they, what they're, what they're holding their experience with what this thing thing is and they can't see the whole picture. I feel like that's the focus struggle in a nutshell is trying to see the whole picture

and figuring out what is okay to ignore. What is the thing that I should really be paying attention to? Another idea underlying all of this to me is the question of technological capability. Like whenever I get in discussions with people about AI, a lot of times they say, well, you know, it's not good enough to write for me. And to me, that is, it's not the right way to look at the question because AI is constantly getting better.

And it is good enough. If it's not already, it will be soon. And I have no doubt that it can write something that I write about better than I can, faster than I can quickly, in the immediate future, if not already. but I really feel like it's my unique voice. That's my expression of humanness is to be able to do that.

And so I want to own it. I think in making your decisions about what's appropriate and not appropriate for you and your relationship with AI, you should not be hanging that on technology. You should be hanging it on values and taste. And you need to really focus on what you're comfortable with regardless of the technology. I 100% agree in the

real important part there is the values and taste that you mentioned at the end. It has to fit in alignment with your values and your personal taste. And that's the part, I think, that makes us a little bit tricky because everyone's going to draw that line a little bit differently. And you'll see other people talking about it online and you'll understand where they drew the lines and that's not where you would draw the lines. Yeah, which is fine, though. I mean, I don't think that there is necessarily a right and wrong answer about this stuff. And, uh, yeah, and, and The technology is changing constantly, which for me has been a real boon on the donkey work piece of this. The, you know, increasingly I don't have to do admin work and that type of things.

And it frees me up to do more of the creative. But that's what I'm monitoring with the technological change is what's, you know, what is the grunt work that I can hand off to this thing? It's not how can I let it replace me. I don't know who came up with the saying of like, you know, I'm so glad AI's here now. It can make art for me, so I have more time to do my laundry. I heard that somewhere, and I thought that was just like the encapsulation of the problem. You know, we need to be doing our art, the AI needs to be doing the laundry.

And that's kind of my overriding philosophy about it all. I agree with that 100%. I think the interesting part about that is the technological capabilities, in making the art. I remember my dad, his dad, my grandpa ran a typesetting business. And when the Mac came out, they were adamant that it's not going to be the same thing. You got to do it by hand. It's going to be better. Yeah, yeah. Right. We kind of know where that went. Now people are making art and making music and all these sorts of things with loops and all these technological tools that were just a pipe dream a long time ago. But you wouldn't, it maybe took a little while,

but people wouldn't say, you know, because you use a graphic design program that you're not creative or because you're using loops, you're not a, you can't create music. And I feel like that's at the heart of this is not really a discussion about AI, but a discussion about technology and how we use it intentionally to do the things that really matter. And then go all the way back to the typewriter, the personal computer. There've been all these different technological revolutions that have changed the way that we work, the way that we make things. I mean, Socrates argued writing made us dumb and less creative. Yeah, it changes the, it changes you.

It changes the process. Socrates was probably right that you're not going to memorize quite as much if you can write it down and look at it later. And there was a statistic. I'm going to botch it. But recently the, like the number of did. that a person can memorize is six or something like that on an average. Like the telephone numbers are often too long.

That's a byproduct of the environment that we live in and we can say, well, that's bad. We should fight against that. Kind of like Socrates did. And I do think that there is a case to be made for fighting for focus. Let's not just get rid of the, it is what it is and jump into the AI slop tide. and our attention spans go down the drain and whatever, like fight to protect some of that stuff. But also it opens up new,

the technology opens up new opportunities and new ways of doing things. And it's constantly changing. I think the minute that you approach it as, okay, I've got it figured out, this is the way to think about this stuff. By the next moment, it is changed and you should be, you should be willing to change your mind again.

Strong opinions lightly held, I think is important if you're going to try to navigate these, the changing waves here. But at the same time, you need to maintain and hold on to your humanness and your expression. Yes. If I were to encapsulate my approach to this, the advice that I would give somebody's trying to figure this stuff out, or even me a couple of years ago, I would say two things. I would consider where AI removing friction is net positive and where removing it is net negative. And then the other thing is I would obsess over quality. And if you are just focused on making the best possible thing that you can, you don't care about taking the shortest path, the easiest path.

You just want to make the best end product. Then I think it'll kind of, you don't need to be too worried about it. You'll figure it out as you go. You'll end up where you need to be. Good advice. Okay, that wraps it up for day. Thanks everybody for listening. Thank you to our sponsors today, Scribe and Squarespace.

If you'd like to be a deep focus listener, go to relay.fm slash focus and sign up. That gets you the ad-free extended version of the show. Today in Deep Focus, Mike's going to be talking about his new virtual assistant. Can't wait to get started. See you next time.