Ash Lamb: Building Your Personal Monopoly! 💡
============================================

by kirupa | Interviews with Creative People: https://www.kirupa.com/podcast/index.htm

Ash Lamb (https://www.ashlamb.com/) lays out a creator strategy built on clear ideas, simple tools, and audience ownership. He explains why distinctiveness comes from combining interests, why video and voice may matter more as text gets commoditized, and why creators should build direct relationships instead of trusting platforms to do the heavy lifting forever.

Watch the interview: https://www.youtube.com/watch?v=zlAGUJH1-Ek
A conversation with Ash Lamb | 1h 16m

ABOUT THIS CONVERSATION

Ash Lamb's work is easy to recognize, and this interview explains why without reducing it to style alone. The central idea is his version of a personal monopoly: a creative identity built not from one perfect specialty, but from an unusual overlap of interests. He talks about illustration, ideas, business thinking, and online communication as parts of the same system. That is what gives the conversation its shape. It is less about how to draw better and more about how to become harder to flatten into something generic.

A big part of that strategy is refusing to overrate tools. Ash repeatedly makes the case that good ideas, honest expression, and recognizable taste matter more than expensive software or elaborate production. That is a useful corrective for anyone who feels stuck waiting for the ideal setup. He is not dismissing craft. He is pointing out that rough work with a live message often reaches people more effectively than polished work that says very little. Simplicity becomes an advantage because it keeps the focus on signal instead of ceremony.

The conversation also gets sharper as it turns toward platforms and AI. Ash knows that social reach can be fragile, especially when work is reposted, abstracted, or boxed in by algorithms. That is why he puts real weight on newsletters and other direct channels. The audience relationship should not live entirely on borrowed ground. His thoughts on AI fit neatly into that same worldview. Execution will keep getting cheaper, but personal judgment, curiosity, and voice still decide whether the work feels alive.

By the end, the episode feels like practical advice for building a durable online presence. Combine your interests boldly, publish with the tools you already have, move toward formats that carry more personality, and protect access to the people who care about your work. There is no grand mythology here, just a thoughtful framework for making creative work that stays recognizable as technology changes around it.

WHAT YOU'LL HEAR ABOUT

- Distinctiveness comes from a rare mix of interests, not just one specialty.
- Strong ideas can outperform polished production when the message feels alive.
- Simple tools are often enough to make meaningful work consistently.
- Direct audience access matters because platforms can change the rules at any time.
- AI can accelerate execution, but it does not replace taste or voice.
- Video and voice may become more valuable as text gets easier to mass-produce.

JUMP TO A TOPIC

- 0:00 - A creator with a clear point of view: Kirupa introduces Ash's minimalist visual style and the larger questions around attention, originality, and online growth. (https://www.youtube.com/watch?v=zlAGUJH1-Ek&t=0s)
- 8:00 - Open-world thinking: Ash connects the freedom of open-world games to the internet as a place where self-directed progress and side quests matter. (https://www.youtube.com/watch?v=zlAGUJH1-Ek&t=480s)
- 16:00 - Why tools are overrated: He argues that clear ideas and authenticity matter more than elaborate production stacks or expensive software. (https://www.youtube.com/watch?v=zlAGUJH1-Ek&t=960s)
- 24:00 - Owning the audience: The conversation turns to reposting, platform dependence, and why creators need direct channels that they control. (https://www.youtube.com/watch?v=zlAGUJH1-Ek&t=1440s)
- 32:00 - AI, text, and personality: Ash shares why he thinks personality-rich formats will matter more as text becomes easier to generate at scale. (https://www.youtube.com/watch?v=zlAGUJH1-Ek&t=1920s)
- 40:00 - Inside the workflow: This section gets practical about his animation, video-making process, and the tools he actually uses. (https://www.youtube.com/watch?v=zlAGUJH1-Ek&t=2400s)
- 48:00 - Reducing production friction: They talk about avoiding theatrics and keeping the creative process light enough to stay consistent. (https://www.youtube.com/watch?v=zlAGUJH1-Ek&t=2880s)
- 56:00 - What becomes valuable next: The discussion shifts to which human skills grow more important as AI handles more execution work. (https://www.youtube.com/watch?v=zlAGUJH1-Ek&t=3360s)
- 1:08:00 - Careful closing reflections: Ash closes by thinking aloud about technology, expression, and the importance of developing ideas with care. (https://www.youtube.com/watch?v=zlAGUJH1-Ek&t=4080s)

TO LEARN MORE

- Ash's Website: https://www.ashlamb.com/
- Ash's Newsletter: https://letter.visualgrowth.com/
- Ash's X / Twitter: https://twitter.com/ash_lmb

TRANSCRIPT

The automatically generated captions have been organized by speaker, lightly edited for clarity, and broken into paragraphs for readability. Names and wording may still contain errors.

0:00 - Kirupa (https://www.youtube.com/watch?v=zlAGUJH1-Ek&t=0s)

Hi, everybody. Today I'm talking with a very talented illustrator, Ash Lamb. There's a very good chance you've seen Ash's creations in your timeline. He's well known for creating minimalistic illustrations that capture interesting ideas and observations about life. You can see some examples on the screen right now of some of his more recent creations. During our chat, we talk about everything from how he got started, the key to marketing online, the power of platforms, extending the idea of open-world video games to real life, artificial intelligence, and a whole lot more. This is going to be a fun chat. You don't want to miss it, so stick around.

Ash, great to chat with you. I've been following your content online for a very long time. I've also been following your content when it's been stolen and reposted by other people for a long time. We'll talk more about that, but first, just introduce yourself.

0:48 - Ash Lamb (https://www.youtube.com/watch?v=zlAGUJH1-Ek&t=48s)

Well, it's a pleasure. Thank you so much for having me. I just want to let the audience know that this is my first proper podcast, so it's an absolute pleasure to do it with you. My name is Ash Lamb. I'm 22 years old, I turned 22 in January, and I've been a content creator for about two years. It was all a coincidence. I'll get into the story later on, but I've been creating visual ideas, especially on Twitter, for about two years: visual ideas on philosophy, business, life, and anything I'm curious about. I create visuals and share them for everyone to see.

1:24 - Kirupa (https://www.youtube.com/watch?v=zlAGUJH1-Ek&t=84s)

So I have to ask: what got you started? How did you begin? At what point did you decide, I'm going to get into computers, get into tech, create visuals, and then share them? I want to hear the full journey.

1:35 - Ash Lamb (https://www.youtube.com/watch?v=zlAGUJH1-Ek&t=95s)

Okay, so it's a pretty interesting story. I was working at a web design agency, and it was my first full-time job. I was feeling a bit lethargic, a bit bored, and I was procrastinating one night because I didn't want to do the work. I was just browsing the internet looking for ideas, and for some reason I decided I wanted to turn my Notion page into a public website, and Notion didn't have that as an option. So I did a bit of research and found this amazing platform called Super. So, and I decided to use it. It went really well, and I built my first website.

Again, it wasn't code, but I felt really proud of myself. I did a bit more research and found out that this platform wasn't built by some large corporation. It was built by someone like me, someone young who liked building things for other people. This person was documenting the journey on Twitter, so he was essentially making the whole thing public for everyone to see, and I decided to create a Twitter account just to follow his work. I didn't have any intention of posting my own stuff, but one thing led to another and I fell down the rabbit hole. I started posting my own ideas without any success.

I got zero likes on average, but I kept going and kept being introduced to other people creating different stuff. And that's when I discovered Jack Butcher. He's British, I think, and he's a visualizer, a very popular visualizer, and he started this whole movement of creating visual ideas. Instead of just sharing text-based tweets, you take an idea you find interesting and turn it into something visual. I bought his course, learned how to do it myself, and the rest is history. It's been a very interesting journey, and what I love most is that it wasn't planned. It wasn't some grand strategy.

It was all curiosity, randomness, and one thing leading to the next.

3:54 - Kirupa (https://www.youtube.com/watch?v=zlAGUJH1-Ek&t=234s)

Yeah, as you'll find, most journeys of that sort are never planned. People often say that when we're building a product, we should build for ourselves, do something we really love, and gradually figure out how to make it apply to a lot of people. But if you start with the idea that you want to build something that's going to attract a large audience, the incentives are a bit misaligned and it can keep you from the destination. Exactly. What was the first visual you remember ever creating?

4:21 - Ash Lamb (https://www.youtube.com/watch?v=zlAGUJH1-Ek&t=261s)

Good question. I've always loved illustration. I've been very creative since a very young age, and I've always made random drawings whenever I had the time. I can't remember what my first visual was, but I do remember that I started by posting sketches on Twitter. They weren't visual ideas yet. They didn't have a meaning or a message behind them. And one day I found this very interesting post by David Perell. He talked about this idea called the personal monopoly, which means combining your interests instead of just focusing on one thing or specializing in one thing. It's saying, okay, I like philosophy, I like business, I like code, I like illustration.

Let's put it all together and see what happens. That's how you stand out from the rest. So I just picked something I found interesting and made a visual, and I fell in love with it because people resonated with it. When there's a message behind it, other people want to tune in, comment, and share it with their friends.

5:31 - Kirupa (https://www.youtube.com/watch?v=zlAGUJH1-Ek&t=331s)

Very true. In terms of what inspired some of the early visuals you drew, what was it? I used to sketch a lot as a kid, and for me it was video games. I used to sketch all the characters. Mega Man was my favorite, and then all the Super Mario characters and so on. So for you, starting illustration, what was the thing you often gravitated toward?

5:51 - Ash Lamb (https://www.youtube.com/watch?v=zlAGUJH1-Ek&t=351s)

That's interesting. I've always been a Nintendo fan, so I'm a huge fan of Zelda. Zelda is my all-time favorite game, and then Mario Kart, obviously, and Mario Bros. Those franchises have always inspired me a lot. I started playing Zelda when I was maybe 10 years old, so it's always been in the back of my mind. Even though my work doesn't really resemble that style, it's always been inspiring to me to put my own ideas into the world and try to create a story, because I always try to include a little character in my visuals. That's what I do differently, because I like to make things a little more playful.

Instead of just creating very elegant visuals, I like to make things more fun and treat it as if it were a game.

6:45 - Kirupa (https://www.youtube.com/watch?v=zlAGUJH1-Ek&t=405s)

On that note, I have to ask you: which Zelda was your favorite one?

6:50 - Ash Lamb (https://www.youtube.com/watch?v=zlAGUJH1-Ek&t=410s)

Ocarina of Time, or...?

6:54 - Kirupa (https://www.youtube.com/watch?v=zlAGUJH1-Ek&t=414s)

Time? Okay, I'm dating. What about you?

6:59 - Ash Lamb (https://www.youtube.com/watch?v=zlAGUJH1-Ek&t=419s)

Link to the past. The link to the past- wow, amazing one, I think Zelda's Awakening on, like, I think, the handheld ones. I think it's a very similar visual style: the isometric top down V, where it's not completely flat but has a slight fake 3D. Okay, yes, yes, I know that. Yeah, yeah, exactly, for watch console was linked to the past. Super Nintendo. This was like Super Nintendo, yeah, 1983 or2 when it came out, and that into an open-ended action RPG game, because I never- I always- play linear games where you just like start off as a clear destination.

That was the first game where I was like, wait, I have options in how I can actually complete my level and that completely opened up a whole new world for me and in many ways that also like had a big influence, like how I approached my career and this life in general as well. It's like there opportunities and open-ended questions that you don't have to answer in the order it's meant to be answered. You can take side quests and just like do other fun things and get back to the main activity at some point later on.

7:54 - Kirupa (https://www.youtube.com/watch?v=zlAGUJH1-Ek&t=474s)

Well, that's a really good point, because I always like to think the greatest game there is the great online game. I used to be pretty addicted to video games when I was younger, and that was until I realized, why play video games if I can just be on social media, growing my own thing and connecting with others? It's almost the same. As you say, you can take all the side quests you want. You can also take big main quests and try to build a big product. There's so much to do that there's no time left to play video games. Yep.

Did you ever play some of the bigger RPGs, like Skyrim or Fallout and...?

8:31 - Ash Lamb (https://www.youtube.com/watch?v=zlAGUJH1-Ek&t=511s)

Oh, definitely. My favorite RPG is The Witcher.

8:36 - Kirupa (https://www.youtube.com/watch?v=zlAGUJH1-Ek&t=516s)

Witcher? Oh, absolutely. Have you played it?

8:41 - Ash Lamb (https://www.youtube.com/watch?v=zlAGUJH1-Ek&t=521s)

I have. That's the one game I played completely, almost 200 or 300 hours, all the way through.

8:49 - Kirupa (https://www.youtube.com/watch?v=zlAGUJH1-Ek&t=529s)

Me too. It's one of those games where the story is absolutely fantastic, gripping, and very open-ended. But the visuals are amazing as well. Every now and then you go over a hill and see the water, the castle, the light coming through. It's almost like a mixture of Lord of the Rings and Zelda, with all the best of the music. It's fantastic. You can even take pictures inside the game. Just imagine how beautiful it is that the developers decided to add that feature. It's so fun. I never played the remastered version, though.

They re-released The Witcher on the more modern consoles, I think both the PS5 and the Xbox Series X, and I heard the graphics were great back then and they made them even nicer now.

9:34 - Ash Lamb (https://www.youtube.com/watch?v=zlAGUJH1-Ek&t=574s)

I haven't tried those yet. My time for gaming has gone down quite a bit.

9:38 - Kirupa (https://www.youtube.com/watch?v=zlAGUJH1-Ek&t=578s)

Exactly. That's actually really cool that you played some of those games as well, because there's something about them that maps to everything we do online. Everyone you track is like an NPC until you actually talk to them, and then you realize, oh, they're real. There's a story arc to them. There's a character path behind all that. The thing that always struck me most about these kinds of games is that so many things we do in everyday life are a game of probability.

In these games, you have skills you can increase, and the higher the skill level, the higher the probability of being successful in a negotiation or some action you're taking. Even today, whenever I encounter something at work or with my side projects, I'm like, okay, my chance of getting the answer I want here is probably 30 percent. What can I do to increase the odds a little bit? That mentality comes 100 percent from games like Fallout and The Witcher, where you have to figure out, okay, I can't deal with this lock right now, or I can't go to this area until I strengthen myself in these other areas first.

So I go finish all the side quests and come back to it, and then I'm like, oh, this is very trivial. I apply that to life all the time, as you mentioned before. That's the beauty of games that aren't linear. Open-world games are so fun. That's why they take ages to complete, because you can almost invent your own story. That's why they're so addictive. I mean, we can talk about games all day and maybe one day do that in real life, but let's get back to the work you've been doing online. When did you realize you were onto something big here?

You mentioned that when you first started posting, you weren't getting a lot of traction, but over time something started clicking. What was that?

11:23 - Ash Lamb (https://www.youtube.com/watch?v=zlAGUJH1-Ek&t=683s)

So here's my secret recipe, as I like to call it, and it's pretty simple. I didn't invent the name, Jack Butcher did, but it's basically permissionless apprenticeship. Instead of talking about yourself and just posting things related to your own story, make a list of people you look up to and try to add value to them in public. That's the key: always in public. Most people would just send a private email to someone. Instead, do it on Twitter, for example, and that way they're much more likely to share it with their audience.

Then their audience will see your stuff, and people will be more interested in you as a byproduct of helping someone else. That's what worked best for me. Again, this wasn't a strategy at the time. It just happened. But because I had so many heroes I looked up to, I liked visualizing their most popular quotes. Most of them saw those visuals and shared them with their audience, and that helped me grow very quickly. I had so much fun because I even got to meet some of them, which was completely mind-blowing to me.

12:39 - Kirupa (https://www.youtube.com/watch?v=zlAGUJH1-Ek&t=759s)

Surreal. And when you were creating them, I'm assuming you put your Twitter handle in the image itself, so even if it was shared without the context, people would know where to go.

12:50 - Ash Lamb (https://www.youtube.com/watch?v=zlAGUJH1-Ek&t=770s)

Exactly, which I didn't, by the way, at the start. I made the mistake of not adding my domain because, again, I was really naive and thought no one would steal my work. But people always find a way to get back to you if they're interested in you. They'll ask others or reverse engineer how the visual was shared. Again, it was great fun. It was surreal for me to visualize someone's quote and then see them like it, comment on it, and share it with their audience. That's how I managed to connect with others.

13:28 - Kirupa (https://www.youtube.com/watch?v=zlAGUJH1-Ek&t=808s)

Actually, finding you took some work. I ended up doing a reverse Google image search on one of the images. There was no comment under it, no website, maybe it was early on or maybe someone had actually removed it. But I was able to find you, and that's how I was like, oh wow, this is really cool work. I've been following you ever since. Your visual style has been very consistent. It's changed a little bit over the years, and I've looked at some earlier images. You know, it changed over time, but you pretty much had a good vision from the very beginning and just stayed with it.

How did that come about?

14:00 - Ash Lamb (https://www.youtube.com/watch?v=zlAGUJH1-Ek&t=840s)

So it is true that it's changed quite a bit throughout the years, because a philosophy I have when it comes to creativity is perfection through iteration, not revision. Instead of trying to come up with a perfect style and then share it with the world, just make tiny iterations every week or every month. Then you can see how your style slowly improves as time goes on. But yes, I love to keep things minimal, simple, straight to the point, and sustainable, because I can't spend hours and hours on every visual. That would be unsustainable in the long term.

So I like to keep it simple and direct so people can absorb the idea much faster. I reduce the number of colors I use. I always use the same line thickness, it's a 14-pixel thickness in Procreate, which is my tool of choice. It works really well to add constraints to your work so you can do things much faster. You don't have to spend hours thinking about each decision, and you can focus on the message and the essence.

15:10 - Kirupa (https://www.youtube.com/watch?v=zlAGUJH1-Ek&t=910s)

What tools do you use? You mentioned Procreate. What's your setup exactly?

15:17 - Ash Lamb (https://www.youtube.com/watch?v=zlAGUJH1-Ek&t=917s)

So I basically use the basic version of Procreate on my iPad, which is a secondhand 11-inch iPad with the Apple Pencil, which is amazing. Then I sometimes edit the visuals afterward on my Mac using Affinity. But that's it. It's pretty simple. There's no complicated setup going on, just pretty straightforward.

15:39 - Kirupa (https://www.youtube.com/watch?v=zlAGUJH1-Ek&t=939s)

Yeah, you share pretty regularly that it doesn't matter what tools you're ultimately using. It's the idea and the output that matter, and there are many ways of creating it.

15:47 - Ash Lamb (https://www.youtube.com/watch?v=zlAGUJH1-Ek&t=947s)

Exactly. I keep getting DMs from people who are just starting out on their journey, asking me what tools I use, because they think the tools make the biggest difference. But that's not true. I've seen people share visuals drawn on a napkin, literally just take a picture with their phone and post it, and they perform better than my visuals because they're more authentic. There's something about them that people resonate with. So there are no excuses. Whatever you have, just do what you can, and you can upgrade your tools as you go.

16:20 - Kirupa (https://www.youtube.com/watch?v=zlAGUJH1-Ek&t=980s)

Yep. Do you ever see a world where, given that you can get visuals out so quickly, AI plays a role? Of course AI is the really big thing these days. Do you ever see a situation where the important, unique thing you add, your personal monopoly, is the idea itself, and the system learns your previous style so you can use AI to create the visuals for you? Do you ever see yourself doing that?

16:48 - Ash Lamb (https://www.youtube.com/watch?v=zlAGUJH1-Ek&t=1008s)

I've tried. I've made a few experiments, and here's the issue with AI in its current form. As far as I'm concerned, it can't reproduce such precise visuals because I spend most of my time thinking about the idea, not so much about how it'll visually look, and AI has a hard time doing that. I use visual metaphors to communicate ideas, and that's pretty tricky to do with AI because you have to be incredibly specific. In fact, it takes longer to write the correct prompts than to make the visual by hand.

It would take ages to tell AI exactly what you want, exactly which element you want in the image, in what way, at what size. It's too precise. So right now it just creates very rough sketches that you can use as reference, but you still have to edit them a lot in order to create something that communicates an idea.

17:55 - Kirupa (https://www.youtube.com/watch?v=zlAGUJH1-Ek&t=1075s)

Yep, that's very true. It's one of those things that keeps evolving over time. What AI could do a year ago, compared to now, has grown leaps and bounds. For example, with video generation, we went from those really bad versions of Will Smith eating spaghetti to much more realistic scenes, like a cat walking through a busy field, and it mostly looks real. So it's one of those things you keep watching to see where the future might ultimately go.

18:34 - Ash Lamb (https://www.youtube.com/watch?v=zlAGUJH1-Ek&t=1114s)

Exactly. I mean, apparently the difference is already quite big. So maybe a few years from now AI can make visual ideas with the click of a button, not just visuals. We'll see.

18:50 - Kirupa (https://www.youtube.com/watch?v=zlAGUJH1-Ek&t=1130s)

So at some point, you were getting a lot of traction and doing really well, and then I think you launched a course that people could sign up for to learn more. How did that come about? When did you decide, okay, now I've hit critical mass, or now I want to share my learnings with a broader audience?

19:08 - Ash Lamb (https://www.youtube.com/watch?v=zlAGUJH1-Ek&t=1148s)

So again, I kept getting emails. I started a newsletter before launching my course, just to go more in depth into my process and whatnot. I kept getting emails from people asking me how I was creating visuals. One day I decided I had no other option. I had to make a course because everyone wanted to know what my process was, and it seemed like a pretty interesting way to monetize my little internet project. So again, I blocked out maybe one or two months to put down everything I knew about creativity and visuals in a Notion document, and I created a reduced version of the course.

I shared it with a few early users, and with their feedback I kept iterating until I had something more extensive. It went really well. People loved the course because it helped them communicate their ideas, even if they weren't looking to become content creators. It helped with presentations, for example. I had many professionals who wanted a way to make more engaging presentations instead of using old, boring PowerPoint slides. They wanted a way to make their physical audience more engaged in the room. Visuals worked well for that. Now I'm working on the second iteration of the course.

I decided to take it down to free up some time to work on the next edition, and I want to make something a bit more condensed. On the first version, I got a bit too obsessed with adding fluff to make it feel like a real course. I thought the longer the better, but that's not always the case.

21:02 - Kirupa (https://www.youtube.com/watch?v=zlAGUJH1-Ek&t=1262s)

Do you charge for the course?

21:06 - Ash Lamb (https://www.youtube.com/watch?v=zlAGUJH1-Ek&t=1266s)

Yes, I did. When it was available, it was around $199, so that was the base price. As for how I came up with that, I can't remember exactly. I think I made a list of courses I'd taken, came up with an average price point, and used that as the first price for the course.

21:31 - Kirupa (https://www.youtube.com/watch?v=zlAGUJH1-Ek&t=1291s)

Yeah, because one of the hardest parts is knowing what to charge for something. You don't want to undercharge yourself, but you also don't want to overcharge. It seems like your approach actually hit all the right spots.

21:43 - Ash Lamb (https://www.youtube.com/watch?v=zlAGUJH1-Ek&t=1303s)

Exactly, and you can always ask your audience. People forget that. You can just ask them what they think. Make it an open question: what would you pay for this, or what would you think would be a fair price? A lot of people will be willing to share their thoughts, and then you can come up with a middle ground, see how it goes, and always change the price. You don't have to stick with it forever. Just because you've come up with a price doesn't mean you have to keep it forever. You can make small iterations and see what works and what doesn't.

Maybe you can add different options, maybe a one-on-one coaching session for people who want to go deeper. There are an unlimited number of options and a lot of inspiration too.

22:29 - Kirupa (https://www.youtube.com/watch?v=zlAGUJH1-Ek&t=1349s)

Very true. With content creation, one thing I've always struggled with, and I see many content creators struggle with as well, is how much content should be posted on a network you don't fully control, like Twitter, X, LinkedIn, or Facebook, and how much should live on your own blog or personal brand, where you know it's going to be there for a while. How do you balance that? Your content is very visual and shared very widely, but you also have it on your own site. Walk me through your thought process there.

23:00 - Ash Lamb (https://www.youtube.com/watch?v=zlAGUJH1-Ek&t=1380s)

Well, that's a good point. I made the mistake of not creating a newsletter earlier, because I didn't even know that was an option when I was starting as a content creator. Then I came across Revue, which doesn't exist anymore. I think Twitter acquired Revue and then shut it down when Elon Musk bought Twitter. Revue helped me take my best fans, or my truest fans, to a place without algorithms. Your account could be suspended any day, even by mistake. You don't have to do anything wrong to get blocked on one of these platforms, which is a big risk and pretty dangerous, especially if you've built a business on top of it.

So it's always good to have a way to communicate with the people who truly care about you in a way that doesn't involve algorithms or strict terms and conditions and all that.

24:07 - Kirupa (https://www.youtube.com/watch?v=zlAGUJH1-Ek&t=1447s)

So true. I used Revue as well, and it was so easy. Before Substack, it was trivial. Before then I used Mailchimp, and Mailchimp was expensive. Once you had a certain number of subscribers, you were paying several hundred a month for basically the ability to write some text and send it. They were one of the only options that could really do it at scale without getting blocked by spam bots and things like that. So when Revue showed up, I was like, wow, this is amazing. I can finally detach myself from this huge, expensive operation for basically free. I don't remember ever paying for Revue, maybe just for the domain.

It was completely free. Then, of course, it went away, but luckily Substack came along and it has import features and things like that.

24:52 - Ash Lamb (https://www.youtube.com/watch?v=zlAGUJH1-Ek&t=1492s)

Exactly. Yeah, Substack is great. Substack works really well.

24:59 - Kirupa (https://www.youtube.com/watch?v=zlAGUJH1-Ek&t=1499s)

Yeah, and it's such an interesting thing, because the incentives for all the major networks are to capture your audience and make it part of their own. That's their number one goal. It's not really, how can I make Ash successful or how can I make Kirupa successful? It's, I want to make them successful because that helps my network be successful. So they control the algorithm. If you put a link in a post, for example, because you want people to explore your website more, they're going to tweak the visibility so fewer people see it. They want to incentivize you to spend more time on the network and not have a link there.

As these networks have gotten more and more prevalent, it almost seems like the way you acquire new traffic is through them. But ultimately you want to figure out how to take the people who are seeing you on, say, LinkedIn and turn them into people who visit your actual website or newsletter, knowing full well that the traditional ways of putting a link directly there will keep the post from being shown widely. So it's like, going back to the game analogy, how do you game the system a bit to make sure you win and the network wins as well? Have you found a way?

Every time I post a link to my website or newsletter, I don't get blocked, but the tweet or post doesn't get any views, maybe a few. It's obviously getting capped by the algorithm. So right now I'm struggling with it. I don't know what approach to take to convince people to take a look at my work on my personal website.

26:34 - Ash Lamb (https://www.youtube.com/watch?v=zlAGUJH1-Ek&t=1594s)

Yeah, I don't have a perfect answer yet. I'm still experimenting and trying to figure it out. One thing I've done recently, especially for the video chats you and I are having right now, is take very small clips, maybe 15 or 20 seconds of what I think is a really interesting part, and post just that directly to the networks. That way people think, oh wow, that's actually an interesting insight. Then they want to dig in more, and at that point the friction of opening the first comment or clicking through is perfectly worthwhile because they're already hooked.

For the long-form writing I do, I actually create a small video of me scrolling through the entire article from top to bottom so people can see it's not just a traditional article. There are visuals and animation and things like that, and that gets people hooked as well. So that's what I've found as a way of trying it out. But I also see other people doing none of that. They just post the link directly and put everything in the first tweet to take people away from Twitter or LinkedIn, and they still get massive engagement. So I'm like, I'm doing something wrong or there's something that needs to be done better.

But 90 percent of my content is aimed at Google search. I aim for the long tail. I'm not writing about the most cutting-edge or latest things, but over time, when people have actual questions, they'll find me as one of the top results for that particular problem. So I'm not very active on social media in that sense because it's time-consuming. All the best advice I've seen says you need to post regularly, daily, multiple times a day for TikTok and things like that. I'm like, I don't have time for that.

I'm going to post when I can, when I can hopefully find value in it, and through Google people will be able to find me. The downside, of course, is that the connection between you, the person, and the content you're creating is completely lost if people are coming through Google. For them it's something very transactional. They're trying to solve something, they go there, spend 15 seconds getting the answer, and leave and never come back. So my bounce rates are very high. Again, I haven't figured out what works, but I've figured out many ways that don't work yet. I feel like I'm eventually narrowing in on a solution.

So you've been focusing on SEO.

28:47 - Kirupa (https://www.youtube.com/watch?v=zlAGUJH1-Ek&t=1727s)

So you've been focusing on SEO, essentially.

28:51 - Ash Lamb (https://www.youtube.com/watch?v=zlAGUJH1-Ek&t=1731s)

Correct, but not in the sense that I'm using tools to see what the top keywords are and things like that. It's more like, what's a problem that I'm having as someone who enjoys this stuff? I just write about that, and I make sure to link to other content I've created. I've never really done a proper SEO analysis. I did look into it a long time ago, and everything that needed to be done as part of it just felt a little gamified to me. I was like, I don't want to do some of these things because it doesn't feel right.

And of course I also work at Google as part of my day job, so I'm like, no, I don't want to do anything that might cause the wrong kind of publicity for me or make my day job complicated. I'm just going to post content, do the best I can, and if it clicks, great. If it doesn't, that's okay. I'm going to keep creating content and win with volume.

29:40 - Kirupa (https://www.youtube.com/watch?v=zlAGUJH1-Ek&t=1780s)

I see. So what do you think about AI then? Because I keep seeing people on Twitter mentioning that SEO is completely dead, that there's no point in focusing on it anymore because chatbots will basically replace Google Search. That isn't true in my case, because I still use Google Search 80 percent of the time. I only use chats when I have a very technical, specific problem that I need to dig into. In most cases, I just prefer searching and seeing what comes up. What do you think?

30:14 - Ash Lamb (https://www.youtube.com/watch?v=zlAGUJH1-Ek&t=1814s)

I think it's. It's a risk, you know, I think for a certain class of things that people use search for today, I think they will find themselves using an AI assistant more. I'll give you an example with some data. Actually, you know the number of keywords that hit Google search console for me. You know, like, let's say, about a year ago was around, like you know, million. You know keywords month. Go through Google and come there. Today it's around two million and that 80% is 100% attributed. You not 100%, but largely attributed to people going.

Like you know, the most common questions are like: how do I make a webcam work in my HTML or how do I round a corner in my canvas? Those are the exact same questions you can ask an AI assistant on. The AI will only give you like a verbal answer. It'll give you the code, snpp and everything directly for you there.

So the most common things that people would search for for technical things on Google before, I think they're increasingly using an AI, which makes sense because the audience they tend to Target is techsavvy enough where they are going to be using an AI assistant for a lot of the work they're doing as well, and so for me, direct traffic to my site has dropped significantly because they're clicking the links through like Google. They're getting similar information actually from my website in the AI results, because if you look at the attribution for it say, like you know, Wikipedia or ccom and things like that.

But at that point though, right it's you're, you're presenting your content in the frame of whatever the AI assistant is. Your logo is missing, the things that make you visually be like, oh, this orange and this like pixelated look and design. That's ca's website. That is completely glossed over. So youve kind of become this very impersonal messenger of information, but the personality that I tend to go for is completely lost. And so I, and so I pivoted more doing videos actually, because I'm like, okay, know that text based content- this point is going to be commodity.

There's going to be no more text based content I'm going to create that will really be unique and it's part of it's going to be a syst b is going to be regurgitated by the AI at some point. It be combined with other sources of information. So what the end user gets will be a higher quality result where it combines my answer and me answer from like, other really smart people and things like that.

So I'm like it's not the right thing for me to go against that, because the developer who's asked this question is getting the best possible results, so it's good for them that I'm- you know that I do this and video isn't something they've been able to replicate just yet. I've seen the transcription demos where you can type in some script, give you some examples of yourself talking, and it creates these versions for you. It's good, but it's misses, missing some of the quirks, and in a lot of my videos actually use like slides and visuals as part of talking to it.

It hasn't been able to mix and match those things effectively yet. I'm saying the word yet because I do absolutely see a world where it's going to be doable, because if they can create all these other things, it's not rocket science to basically take me talking and have like some slides and visuals created and how this all work. But you know it's a constant, you know battle to chase after, like you know, make sure technology isn't completely catch up to you, and so that's what I've been doing. And of course I also publish books. You know I work with a publish content and there's a large enough audience worldwide there.

I know often get emails from people going. Like you know, I'm in this library and I saw this book and I learned about it. It's actually really cool. So there's a world that will not be, you know, using the internet and Ai and all these things, at least early on, and so I still have that outlet right now. But if you ask me five years from now, I think it'll be a very different answer, almost to the point. I'm like you know what I'm basically just doing live sessions from like teaching people in real life, because that hasn't been replaced with AI just yet.

I hope get there, but it very well, might well.

33:30 - Kirupa (https://www.youtube.com/watch?v=zlAGUJH1-Ek&t=2010s)

It's a very uncertain future. Very interesting. I think about it a lot. I even lose sleep over it. I'm curious, do you post those videos on YouTube, or where do you post them?

33:42 - Ash Lamb (https://www.youtube.com/watch?v=zlAGUJH1-Ek&t=2022s)

I post them on YouTube. That's where they go, because long-form works there. I've also built up a network on YouTube over the years, so that's home base. Then I post short five- or ten-second versions on Twitter. I use a link on Twitter and you get the thumbnail preview below, but of course the algorithm is like, no, just upload the videos directly. So I upload the videos as well. I've gotten to a model where I actually record my videos in this PowerPoint format where I have two halves of the screen.

The reason I've done it that way is that it makes it easy for me to repurpose them into TikTok videos, where everything is vertical instead of horizontal, without any modification. At some point I'll talk about how I do that in the future. Otherwise I'm spending a lot of time thinking, okay, Instagram needs to be square, LinkedIn needs to be widescreen. YouTube, if it's under a minute, needs to be vertical, and TikTok needs to be vertical. I'm like, I don't have time for this. My goal is to educate people and do all these things.

I also like to learn these tools because that helps me empathize with creators, and I build tools for a living in my day job as well. So I like to learn how decisions are made, why people do the things they do, and why tools are built the way they are. I didn't want to outsource that or defer it to someone else, even though they'd probably do a much better job than me. So I've been tweaking all these ways of making my video creation as efficient as possible.

35:01 - Kirupa (https://www.youtube.com/watch?v=zlAGUJH1-Ek&t=2101s)

Well, that's so interesting. It's easy to tell you're an engineer.

35:06 - Ash Lamb (https://www.youtube.com/watch?v=zlAGUJH1-Ek&t=2106s)

Oh yeah, unfortunately. I'm basically an engineer of video games, so it really is the best of both worlds. The two key words are optimization and efficiency.

35:18 - Kirupa (https://www.youtube.com/watch?v=zlAGUJH1-Ek&t=2118s)

Exactly. When I was much younger and had no obligations, I had all the time in the world to fiddle with this stuff. Right now my time is very constrained. I have a family and all these things, and I don't want to miss out on that. So where I can optimize is in this other work, because I don't want to optimize the time I spend with my wife, daughter, family members, and friends. That's the thing that's going to last forever. But I can absolutely be like, you know what, that TikTok video is okay. It's not zoomed in exactly right. It's pushing a larger version onto a smaller screen.

If you really want to look at a coding screencast, go to YouTube. But if you're on TikTok and still watching it, something is interesting to you, so just go ahead. I'm basically sacrificing a bit of quality for speed, kind of like a bad image compression algorithm. It's like, yeah, JPEG is fine. PNG might be better, but 80 percent JPEG, out the door, done, don't even talk about it. Well, it's worth it in the end, right? Because you get to spend more time with your loved ones doing what you love, so that's what matters at the end of the day.

Especially if all this content is going to get commoditized, there's no point spending thousands of hours obsessing over a zoom setting or something. You also see the worry in Hollywood and the movie studios. I think one of the more recent contract negotiations and strikes was really driven by the use of AI in making visuals and all these things work. I watched Dune 2 recently, and it's a fantastic movie, absolutely amazing. But then I'm also like, how long will it be before AI can create some of these scenes? Yes, the director is extremely talented, so AI can't create some of those shots exactly the way they were done.

But at some point AI is all about training. It trains on legacy data and then creates something based on that world. Unless you're really pushing the envelope, which very few directors probably are, most are trying to create great content using traditional techniques. That's the kind of world where AI is always going to be like, yeah, this is fine. If you're using a style that somebody else came up with a couple of decades ago, in terms of how you pan and zoom and so on, it will absolutely create a video in that exact style.

So the question becomes, how much time do you spend staying on the cutting edge, one step ahead of the AI, before you just burn out?

37:40 - Ash Lamb (https://www.youtube.com/watch?v=zlAGUJH1-Ek&t=2260s)

Well, I think the key is not to compete against AI. That's what I've realized. Instead, focus on your own personal stories and double down on your personality. In your case, you're obviously a really funny person. You have really good humor and a very obvious personality. So instead of writing really technical articles for people trying to solve a very specific problem, double down on your way of communicating things, like Ali Abdaal does. He started by sharing very transactional or informational content, and people discovered him because of the problems he was solving. But people stayed because they liked how he talked and how he communicated his ideas.

It doesn't matter what the videos are about, people will watch them because they resonate with who he is. I don't think AI can compete with that just yet, or anytime soon, because that's human nature. So that's what I want to focus on this year: creating a deeper connection with my audience and with anyone who likes my stuff. I'm still thinking about how to do it, but I think that's what I'll double down on.

38:58 - Kirupa (https://www.youtube.com/watch?v=zlAGUJH1-Ek&t=2338s)

Yeah, that's a really good path to pursue. I noticed that you also do YouTube videos. I think you posted your first video this last month. How did that come about? What tools did you use? How did that whole thing happen?

39:11 - Ash Lamb (https://www.youtube.com/watch?v=zlAGUJH1-Ek&t=2351s)

Well, first and foremost, that took ages to put together. I obviously underestimated how much time it would take me to create a video at that level, because I animated each frame by hand. I'm a perfectionist, and it took quite a lot of time. I used Procreate to animate each frame traditionally, and I also used DaVinci Resolve, the free version, to stitch everything together. Then I just recorded a voice-over, and it was great fun. It took a lot of work, so it's unsustainable. I'm trying to find a much better way to do it now, but oh my goodness, did I have fun.

It was so rewarding, and I got a chance to go deeper. I'll definitely be showing my face in the next videos and experimenting with different ways of doing things. Showing your face makes it more personal.

40:11 - Kirupa (https://www.youtube.com/watch?v=zlAGUJH1-Ek&t=2411s)

That's exactly it as well. So how did you make the squiggly lines actually look like they were waving in every frame? Did you do that manually, or was there a tool that did it?

40:20 - Ash Lamb (https://www.youtube.com/watch?v=zlAGUJH1-Ek&t=2420s)

No, I used the effect in Resolve for that, so that was helpful. I like that you noticed it. I was so happy when I discovered that DaVinci Resolve had this feature. It's just an effect you can create using nodes. Connecting them together took me quite a bit of time to figure out, but now I have this way of making things look fancier than they really are. Everything else was frame by frame, and if you pair a few animations with this effect, it looks like the whole thing is animated by hand. That was a sneaky trick I used.

40:56 - Kirupa (https://www.youtube.com/watch?v=zlAGUJH1-Ek&t=2456s)

That's actually really cool. I also like how you described building nodes and connecting them to create the output. I've worked in developer and design tools for my entire career. I started with Flash many years ago and then moved into creating developer and design tools. I've always found it fascinating to see how tools can help people be more creative and build things in a scalable way, so the result works well when it's handed to a developer. That scalability matters because creative work often passes between designers and developers; the representation needs to remain understandable as the artifact moves from one person to another.

I like how naturally you said, “I'm just going to use nodes and make these things work.” There's an old saying in the tools world—perhaps only I and 10 other people talk about it—that you want to solve a problem in the domain where it exists. If you have a visual problem, you want to solve it visually. Suppose you want to draw a square or a circle. You don't want to write code or letters to produce a square. You want to select a pencil icon or whatever is appropriate and simply draw the square or circle.

Over the years, we've found that when people modify an object's properties, especially in a cause-and-effect relationship, node-based drawing resonates with them more naturally. The most basic example comes from electronic music. You often have amps, mixers, and other devices connected by a series of wires. It naturally makes sense. You may think, “Wait, I'm making music,” but something in our brains makes this cause-and-effect relationship represented by boxes and lines feel natural. You see it in many other cases too. Many 3D tools, animation tools, and rigging tools use the same line-and-box concept to make things work visually. It's fascinating to see that approach persist.

People said that once tools improved, nobody would work this way anymore. AutoCAD and similar tools were doing some of these things 30 years ago, yet the approach is still used today. It's always great to see that style of problem-solving and tooling continue. Despite decades of changes in computing power and interface design, that same visual model continues to match how people understand dependencies.

42:51 - Ash Lamb (https://www.youtube.com/watch?v=zlAGUJH1-Ek&t=2571s)

Well, even website builders, right? Back then you had to use code to build a simple HTML layout, and now you can just use Framer, for example, which is what I use to build my websites. You can do it 10x faster and have much more fun in the process. You don't have to spend hours wrestling with HTML and CSS just to create a very basic layout.

43:14 - Kirupa (https://www.youtube.com/watch?v=zlAGUJH1-Ek&t=2594s)

Exactly right. If I had to pick one field that AI is going to completely turn upside down, it really is web development. The reason I say that is, back in the day, this is how you built websites: you used Dreamweaver or FrontPage and did it visually. Then at some point it became, no, you must learn all the internal details of how everything needs to be done, and that made the whole world really complicated just to get Hello World on the screen. You spend a couple of days learning all these weird extra things, and now we're back to a model where you know what a web page is.

It's basically text on a page. It's visuals. It's something I can easily see and make sense of. There's no reason it needs to be done in all these very complicated ways. The abstractions will truly be handled by AI and tools. That's a really good point, and it makes me think, as you already know, I've been trying to learn the basics of code in the past few months and I've been having a lot of fun. But there's this voice in my head that keeps telling me I'm wasting my time and should focus on something else.

I don't know if I should find a middle ground or if maybe this is the best time ever to learn how to code, because if you know how to code, you can leverage AI much better. That's also true, right? Because you have to know what to ask in order to get a decent result. So what do you think? Because you're a complete expert and I'd love to hear your thoughts. Is it worth it? Am I wasting my time?

44:44 - Ash Lamb (https://www.youtube.com/watch?v=zlAGUJH1-Ek&t=2684s)

I mean, it goes back in one of two ways: right, like, let's use an example of like you know, you want to paint a really know Great Masterpiece on like canas, for example. Right, some schools, they'll tell you to do that first: get the wood, buy the canvas, stretch the canvas yourself, mix, learn how to mix the paints, buy the basic powders, mix it together at the ratio right, and all these things, and that'll give you more fundamental understanding of what you ultimately want to do. Your goal, of course, is to create like something. You. You have an idea in your head. You want to visualize it.

You want to go from like zero to an idea, to something that I can show people and like. That's my destination. I want to share what I created with the world. That's the fastest path. I want to get there. Now. The question is: can you get there? Can you show people something that you know, that truly represents what you're trying to create, without having to learn all these fundamental steps and things like that?

If you're someone in the who's in the business of selling the wood, selling the canvas or making the paint mixes and so on, you're going to be like: no, you absolutely need to know this, you need to know all of these things. When I look at web development, I find that that group of individuals dominates so much of the conversation. You know they always are like the ones who work for, like the. You know the Big Tool makers, or the framework makers, or the nitt details are like you must learn react, you must learn view, you must learn angle, you must all these things.

For some people you might need to know that because what you're creating needs to require that level thing for 90% of the world. You don't. You just want to go to an art store, buy the canvas the way it is already made. You just buy the brushes, buy the. You know acrylics or oil painting. You know acrylics are great because you know they dry quickly and it affect is mostly pretty good in many ways as well. Just do that and you have an idea can share with people. And most of the world will not know the difference between that ask: make that painting by stretching the canvas on its own.

Or you go to store and just make the Cannabis. They're going to be mesmerized, create. And so, from that point of view, when you look at it entirely from like, who am I building it for and will they care? 90% Of the answer is no. Do the shortest level of steps you need to take to make them happy, which, in the case of websites and web development, you don't need to go and do all these things. Like? I've been blogging now for like almost, you know, since 1998, so, almost, like you know, 26 years.

At this point, I use dream weer, even today, and people like really, I'm like I used dream weer and I used FTP and I used Cloud flare for my CD and I'm like, yeah, it works. You know, while you're spending like six months learning about this latest technology, that I spend those six months creating content and creating all these things and no one knows, like no one cares. When they visit your website, going like exactly built it, you know it's like wait, they're like wow, it's really fast, loads really Snappy, works on my mobile device, it's accessible. I'm like great, I hit all the check marks how I built it.

It's entirely up to me on how I did that exactly. Imagine a use refusing to use the website because it hasn't built on the latest framework, right? It doesn't make sense. No one cares exactly, and so that's why I look at you know, like you mentioned earlier. Right, AI is a tool. It's just going to help you know, speed up your process and how you get your ideas through. That's exactly how I view everything that we do in general as well. It's like I might have an incentive to say: you know you must use tools a, b and c and then have this approach.

But if I take a step back, though, that's just my incentive, you know, the end result for the user is not going to be very valuable. So it really about like the problem you're trying to solve and I think for 90% of people, maybe even 100% of people, probably, like you know, a year from now, all you need is like maybe a newsletter, maybe like WordPress, you know, still very, very great for like being able to create content, have a theme.

You can customize it, personalize it, want to the less amount of time you can spend on the, on the mechanics of getting something done, as opposed to getting something done. I think that's the really the best way. Like I call it theatrics, you know, like a classic thing of like you know especially large organizations and companies. You worked as a web developer for an agency and things like that. You know it's like you know, some of the work you do is just Optics.

It's just to kind of give you the illusion that you're really doing work, but the reality is that there's no real purpose, doesn't adjust the quality of your output by you making you know this approach or not. It's just Dogma that teams and people follow, and so I've been very much on a. You know how can I make sure I don't fall into the Trap of mistaking the Optics of doing something with actually doing something?

And it's so easy to get into that trap at some point where you're like: oh, I spent six months creating hello world in, you know, in react when the reality was like you never need to do it in the first place, because that's not why you started doing this. You, if you want to learn it, great, but if your goal was to build a website or if you want to build something to share with people, you probably spent way too much time on something that didn't move you closer to that goal. That's interesting and I agree.

However, as long as you enter the realm of web apps- not just websites with content- I think it's beneficial to understand the fundamentals, because then the difference is exponential when you use a, because if you understand how things work- and you don't have to be an expert- just learn the basics- maybe go through the auding project right and just go through the basics, then you can ask chat GTP much, much better questions and you can create proper apps, not just content app- content apps that actually work and scale. So that's where I'm at right now, because I've been thinking of building.

A few ideas I've had, but I don't know enough to leverage AI to build them right. So I know plenty about creating content websites. That's amazing, and it has been completely democratized for most people. However, as soon as you want to build an app, I think you need to have some level of understanding to build something decent.

Absolutely true, you know it's like you don't know what you don't know sometimes, and the more you know about the behind the scenes details, like you mentioned, you'll be able to more precisely use the right words and right choices, because it's always very subtle, right and it's kind of like exactly from like you know something that shows, like you know, 50% Mastery of a topic to 80% Mastery. Just, it's only 30%.

It's a number, but the reality might be like it might be like two years exactly to get that point, but then you can quickly spot someone who has that extra 30% of knowledge, because when you talk to them you get that sense of like. Okay, they know they can jump back and forth in, you know, in time and in depth and breadth of the topic to be able to quickly get you to the answer you're trying to get to and even foresee where you might be going next. So the solution they providing it has has like experience baked into it behind the scenes.

So, yeah, very true, we'll see where all this goes. You know it's all a matter of experimenting, having fun. We're so early in the process, right, and we're only getting start when it comes to AI. You know there's so much to do, so much to see, yeah, but I still do hold the hope that you know, even for building like a very heavy app, you don't need to know all those details like all you like. You know here's like, let's say, a slide deck of like screens that my app needs to have.

Or here's I'm going to write, I'm going to verbally describe my application like a traditional spec or a product requirements document, PRD- and I'm going to give it to my AI assistant. He's going to create the entire website for me, with all the interactions and things like that.

I do think we'll get there, you know, probably sooner rather than not, and because I'm seeing like early signals of like people being able to build single page websites with just like a screenshot of a of their design or like a sketch on a napkin, and using AI to make this work, and they look at tools like you know, both like Microsoft power apps, which are using some AI capabilities on. You see, builder. I has a lot of tools for like building, like you know, like apps based on a template, but using AI to extrapolate large of details for it.

I think there's going to be a world where you know what we call Professional developers, the ones who know how to make the canvas and, to you know, tie the wood together and so on, and then we have like low code and no code, like your framers and like your web flows of the world that's going to melt together because the end result really is: can you look at the out put and be able to be like: okay, I told my AI to create something for me.

It created 10 versions of what I think what it thinks my actual, you know- idea of the final desent should be: do I have enough context to be able to choose which of those 10 I like, which of these 10 I do not like, and then use that to help the AI refine it further and further? In many cases, at this point, you become more like a creative director, like an art director, where you have aam of people pitching you ideas.

You know enough about the industry or the context to be like: this a good idea, this a bad idea, but that doesn't maybe always require to know, like what kind of pen do they use for making the art, or what is the thickness of the of the board they printed this on. You need to know all that. The bigger question is: will this, will my client like it, will this sell? And things like that which might be an art director kind of a thing? Every domain has their own set of like you know areas where they're going to be focusing on.

You know, like I always imagine in my case, right, I would love people to be like. You know telling AI, like I want to teach people this complicated way of sorting numbers and I want to do so in a way that's humorous, includes visuals, use my old visual style to, like you know, learn from that. And then it creates like 10 variations of a document for me and I'm like: okay, I like this, I don't like this. I like the intro in this one, I don't like the way you describe the body part of this one: too much fluffier, too much humor, unnecessary.

And then it goes back, uses my feedback to create round two and round three and round four. At that point I do think that the output, based on how I prompted it, how I interpret it, would be exactly what I wanted to do with me actually having written a single word or drawn a single Pixel without actually having, you know, Outside The Prompt that I gave it. I do think programming will get to that world. And so, like I always think about, like, Should students learn computer science today or should they go into, like you know, learning really deep technical skills? Do they go into it?

My hunch is like no, probably not, because as long as you know what makes a good app and what doesn't make a good app, usually based on output, like you know, is it performant? Is it look good on my phone? User experience, user experience- like that's what really matters. How it gets built, at that point I don't care. You know, I don't care if the AI is using like react or if it's using, like you know, some Library, like you know like motion or frame or you know whatever.

I don't really care that point, because all I really care about is: I have a problem, my customers have a problem, my users have a problem. Is my app helping you to solve that? This goes back to what we talked about earlier is like: will people really care how you create your visuals or like how you create some of these things, as long as it the message very, very effectively and so yeah. So I guess, to answer your question, from like a l, a roundabout, you know view, it's like: no, I don't.

It's good for you to learn it after you've gotten to the point where, like I have a product in Market, I have some breathing room, I just want I'm curious to know, like, how was some of these things done? Like I like to learn about, like older programming languages, just because I'm like, oh, I'm curious, see what it does. But I wouldn't make that my first step as part of what I'm doing. It's mostly a curiosity thing of like I just want to get deeper into what I'm already doing, but it wouldn't be at the expense of not having shipped anything, not having solved the main problems here to solve.

55:26 - Kirupa (https://www.youtube.com/watch?v=zlAGUJH1-Ek&t=3326s)

Yeah, that's a very good point. And remember, no-code is still someone else's code, so somebody will have to maintain it. We'll still need new computer scientists to do that, don't you think?

55:43 - Ash Lamb (https://www.youtube.com/watch?v=zlAGUJH1-Ek&t=3343s)

Absolutely. But if you look at the number of people you need for that, though, it becomes very, very different. But here's the thing. Right, let's say that. Let's say the market is, you know, this large. I'm going to make it a screen so it actually shows up here. And in order to build a website like, let's say, you know, I want to create like ash lamcom, you know, in a world before we had low code and no code and so on, it required me to understand HTML, CSS and JavaScript, and I understand, like, where do I host it, how do I make sure it's performant, and things like that.

And then, after that, I need to learn about the visual skills that you have, like the color scheme, the rounded corners, the arrangement of things, and, like, the style of presenting things, like testimonials. So it requires shifting from a very technical, mechanical world to more of a creative world. And then it's the content itself. Like, how do I write the words in a way that makes sense? How do I get people, you know, hooked on the tagline before they go further into it?

Like, how do I style the last word in the sentence so you gets extra emphasis like that's where the art and creativity comes in. That you know you can't really teach people, you just have to learn it through like your indirect ways of like your history and art and sketching and having observed great designers and great artists and things like that. So it requires you to learn a huge set of like skills over a period of time that you know it takes probably you know, five, six, 10 years, maybe a lifetime, for some people to ultimately get get better at.

But if now you're able to say, like you know what, I can bypass all this HTML, CSS and JS jargon, I can bu going to bypass all that. I'm even going to bypass like the copyrighting angle of it, because the AI does a really good job, like you know, taking, like you know, look at, like what David ogleby did for like marketing, like in the 50 example, like you know, learn from his style of marketing and use that in making this work.

And so now what we've done is like we've taken jobs and roles that required like four or five, three or four people who had went to college, went to like degrees and so on, to be like, nope, it's being done by AI altogether. And so now you're basically the point where, like, the only the real big value you're adding is the, your uniqueness. You know what makes you tick like you, your combination of philosophy and art and all of these things into like, your final output. That's something that AI can't really create, at least, hopefully not for a very long time, and you know so.

That's the part you're focusing on. Everything is automatically abstracted away for you, and so that means the number of people you needed before to like make this work has gone down significantly. The time we have spent on learning about this has gone down significantly as well, and so you will need computer scientists, but you only need computer scientists- those who genuinely care about creating Frameworks and creating these tools, because they're like, oh I love waking up every morning, I love creating tools. That's me. I love creating tools. I love that's you and things like that. Great, but there aren't too many of me out there I can count on like.

Maybe, like you know, I probably about 50 people in the industry. I know of like who have that. It was probably a lot larger, like you know, eight, nine years ago when I first got in the industry like, like you know, Adobe and Microsoft and IBM and Autodesk and so on. I was like, okay, there's a lot of people who are doing tooling and things like that, you know, as part of their day-to-day job. Now it's like a smaller and smaller number, and it's not because people don't care about tooling anymore, it's just that there's not an opportunity anymore where you can use that to basically make things work.

We commoditized around like BS code and like you know, all these various areas and like it's a great place to be, and so that's what's going to happen to computer science as well. You will still need computer scientists, but you won't need them in the number you needed before to be able to create the kind of things needed, because those jobs will no longer be as relevant as they were. You might have them. Will you be able to command the same amount of- like you know- salary and like your quality of life and like that? Probably not.

And this is going to be kind of like the old example of you know, in the 1700s there were a lot of people who made Horseshoes and, like you know, blacksmith and so on, and then, as cars became more popular, as things became more resilient and we had like carbon fiber and things like that. You didn't really need as much smithing work as much as you might have needed to before, and so you had. Creative destruction is a classic term for like people who went from like an older task to being able to do something new.

My only worry with AI specifically is: okay, you know, the idea is that you now you replace, like, lower value tasks with higher value tasks, and it's a constantly shifting scale, but with AI replaces some of these things. What is the higher value task? Because the higher value task itself- what we've for told is like going to be Irreplaceable for a long time- is being replaced very rapidly. The kind of creative thinking, like legal, like law professionals and accountants and act and things like that. I'm like: okay, so what do we do with all these individuals like, what is the transition?

People who were like know blacksmithing, they can become machinist and tooling specials and so on. You know, the jump was high, but there was a path here. But you take someone who's like a computer scientist. What is your next step? Where would they go after that? What skills they have? They learned that'll then make them be effective, like, oh, they're math, you know, they know mathematics. They can like to. You know use mathematics. I'm like, well, a can do that as well.

Like, I can ask any question about, like, modeling the growth of the bird population in North America when, you know, the weather is X, Y, and Z. It'll give me that answer very quickly and show me the work, and I'm like, that's actually reasonably well done. My worry is, where does creative destruction take us when there isn't anything beyond? What exactly do we do that's going to command a quality of life that's hopefully better? You know, machinists made more than blacksmiths. Blacksmiths made more than the people who were, like, digging in the mines too.

I not car at the or, whereas I don't know what's going to be like. What is the next big thing that we can all work on? Because the world isn't need an infinite prompt Engineers, because prompt engineering brings a skill set back down to anybody who has an idea and a way of representing it. They can be effective in it. So yeah, that's my, that's the worry. That's what keeps me up at night.

1:01:10 - Kirupa (https://www.youtube.com/watch?v=zlAGUJH1-Ek&t=3670s)

Well, the next step is enlightenment. Exactly, right? People talk about universal basic income and things like that. I don't know where it's going to land, and I think we need to figure it out quickly. If you had told people five years ago that AI would create full-motion 4K video that looks almost real, they'd have said you were being way too alarmist. But given the speed at which things are moving, this is something we need an answer to sooner. Oh, it's fascinating.

1:01:39 - Ash Lamb (https://www.youtube.com/watch?v=zlAGUJH1-Ek&t=3699s)

Yeah, I saw Sora last month and I was blown away. Those videos look completely realistic, and they'll definitely replace stock footage as far as I'm concerned, 100 percent. I stopped relying on stock libraries as much because I can get things like rough assets from an AI assistant, modify them on my own, and use that. There's a lot to think about. It's a bit sad in some ways, especially as a creative, because I love creating for the sake of creating. I'm not looking for ways to optimize my process, so we'll see where this leads.

1:02:18 - Kirupa (https://www.youtube.com/watch?v=zlAGUJH1-Ek&t=3738s)

Yeah. Also, one last thing before we start wrapping up: for your videos, I used to never have background music looping because I didn't know which sound library I should use or how to create it, and the ones I wanted were way too expensive. Then I found this AI music generator by Google. I think it was called something like the AI music lab. You describe what you want, like a basic drum beat or an ambient space loop, and it creates eight versions of it. There's even a little checkbox asking if you want it to be loopable.

I'm like, yes, make it loopable, so it adjusts the beginning and end to make it work. It's been so convenient. So now every video has a different, very quiet loop in the background, maybe at 30 percent volume. I was able to do that using AI because I have basically no idea how to create music on a computer. I play the piano, but I don't know what MIDI tools to use or any of that. There's going to be a world where people don't know coding, don't know how to create visuals, and don't know how to use Photoshop, Procreate, Illustrator, or Figma. They're just going to use AI.

Just like I was blown away that I now have music in my videos, they're going to have visuals in their content as well. A professional musician would probably say, 'Wow, that loop is low quality. That's terrible music.'

1:03:37 - Ash Lamb (https://www.youtube.com/watch?v=zlAGUJH1-Ek&t=3817s)

Yeah, true. But to me, someone who has no idea what is or isn't good, it solved the problem. One of the epiphanies I had was that we're going to see more and more of this. My daughter is going to be six years old, and she might grow up in a world where she's able to create things without realizing the details that go into creating them. That's the really exciting part of AI as well. I'm talking a bit doom and gloom here, but the exciting part is that more people will be able to be part of what we call the creator economy and do really good things.

The question, of course, is how existing creators adapt, and whether the economy is large enough to support that.

1:04:13 - Kirupa (https://www.youtube.com/watch?v=zlAGUJH1-Ek&t=3853s)

Exactly. And will the content be valuable enough by then? If everyone can do it, the value diminishes. It's the whole supply-and-demand thing, right?

1:04:25 - Ash Lamb (https://www.youtube.com/watch?v=zlAGUJH1-Ek&t=3865s)

For a long time there was always demand, but supply was limited because few people knew the magic tricks to get everything up and running. In a world where supply becomes infinite, demand doesn't scale the same way. There are only so many people in the world, and the population is even shrinking in many countries. So how do you maintain a certain level of quality and sustain a living for all the content creators when it becomes a commodity? Everyone will be a media company.

1:04:54 - Kirupa (https://www.youtube.com/watch?v=zlAGUJH1-Ek&t=3894s)

Right, a walking media company. It's going to be mind-blowing. In fact, Netflix won't even exist because people will create their own movies on the spot. Did you ever watch, I think it was called Bandersnatch, that interactive Netflix movie where the movie plays, you get questions, and depending on the answer you choose, the scene changes completely?

1:05:21 - Ash Lamb (https://www.youtube.com/watch?v=zlAGUJH1-Ek&t=3921s)

I haven't, but it reminds me of The Witcher. Exactly. It's basically a live-action film where your choices change what happens. But imagine that becoming dynamic. Before, you needed people on a set recording 80 different variations. In the future it'll be generated on the fly. For me, the thing that's most exciting is that I love gritty TV like The Wire or The Sopranos. Those shows have ended. But imagine AI letting me say, give me two more seasons of The Sopranos, and it creates new seasons of a show that already ended.

Some of the actors have even passed away, like James Gandolfini, and yet it could create new episodes that look and feel exactly the same. I feel bad for the creators, like David Chase, who created The Sopranos. But as a consumer, I'm like, this is perfect. And my version of The Sopranos would be very different from your version because it would optimize for our different interests and tastes. So we're creating this really interesting world where personalization gets taken to an almost unnatural extreme.

1:06:28 - Kirupa (https://www.youtube.com/watch?v=zlAGUJH1-Ek&t=3988s)

It's crazy. As they say, reality is always crazier than science fiction, right? So we're about to experience something completely different, almost like Ready Player One, the movie and the book, at some point. I can imagine a world where we have the option to immerse ourselves in this VR or AR reality and create our own world on the spot, and then nobody wants to leave because why live in the real world if you can just be in a simulation? Maybe we are in a simulation anyway. Exactly, right. Well, Ash, it's been great chatting with you.

We started off talking about your work and then drifted into AI, but it is the elephant in the room. So any parting thoughts? Anything you want the audience to know before we end?

1:07:42 - Ash Lamb (https://www.youtube.com/watch?v=zlAGUJH1-Ek&t=4062s)

Well, two things. First, I'm going to be writing a newsletter issue, maybe not long-form, but definitely something about AI, because I want to synthesize everything I've learned. And you should too. I don't know if you've already written a proper long-form post about AI, but I'd love to read one from you.

1:08:01 - Kirupa (https://www.youtube.com/watch?v=zlAGUJH1-Ek&t=4081s)

You know, I want to write it. The thing is, this goes deeper into my style of creating content, because my day job involves living at the intersection of tools and AI and bringing to reality a lot of the things I talk about. I'm always trying not to muddy the waters too much, where I might have an opinion that goes very much against what Google as a whole might have, or maybe they agree with it too, and I'm like, I have so many other things I could talk about instead and not create conflict or drama. Drama always seems to find me.

So AI is one of those topics I'll talk about in a podcast, in a more casual form, or in person. It might be a while before I write about it in a more formal way, mostly because I don't want to add to the existing narrative on AI that companies spend a lot of time crafting. But it's always in the back of my mind. Exactly. It's a fun constraint, like you mentioned. A lot of people talk about things that relate to their jobs, but that can also create unnecessary complications that take time away from the things I do care about.

I enjoy the parts of my day job that aren't related to AI specifically, so why muddy the waters if I don't have to? That's really the way I look at it. But I do look forward to reading your piece when it's out.

1:09:26 - Ash Lamb (https://www.youtube.com/watch?v=zlAGUJH1-Ek&t=4166s)

I think it'd be contrarian, I'll tell you that, because I have a few ideas lingering around. The nice thing about contrarian takes these days is that you know AI probably didn't write them, because it's trained on 80 million versions of the same content. So when something sounds completely different, I'm like, ah, that's a unique take. But I'm also waiting for the moment when AI can say, write a contrarian take on the future of AI, and do that too. The speed has been rapid, much faster than I ever imagined.

1:10:01 - Kirupa (https://www.youtube.com/watch?v=zlAGUJH1-Ek&t=4201s)

Really? Do you think so?

1:10:02 - Ash Lamb (https://www.youtube.com/watch?v=zlAGUJH1-Ek&t=4202s)

Yeah. Where we are today, I thought we'd maybe be two years from now. The fact that we're already here is wild.

1:10:10 - Kirupa (https://www.youtube.com/watch?v=zlAGUJH1-Ek&t=4210s)

Mind-blowing. So you're a technologist. How long have you been thinking about AI?

1:10:16 - Ash Lamb (https://www.youtube.com/watch?v=zlAGUJH1-Ek&t=4216s)

I'd say for almost 10 years. Wow, amazing, ever since. You know. This goes back to video games again, right, I think it was Skyrim. Skyrim, like almost 10, 12 years ago, started randomly generating Quests for you. I think they called the radiant Quest system. Oh, is that true? Wow, right. So 10 years ago, yeah, almost like. When did Skyrim even come out? Was it like 12 years, 13 years now? Basically, yeah, probably on the time frame, right? Yeah. So I was like: wait, why this Quest? The same, I'm going to a cave. It looks different, but it keep collecting a similar set of things over and over again.

It never seems to end. And when I went online, I realized, to make the game last longer and cases, these quests are randomly generated, the worlds are randomly generated, the, what you do and the people you talk to, dialogue, randomly generate. I'm like, wow, this is a. I never expected this. And then, when I started digging deeper into it, that's when I started realizing, okay, there's going to be a world where AI is going to be more effective in doing some of these things. We're not there yet. You know we had chat Bots forever, right? You know, go to a support site.

You know you ask a chat a question and try to like give you some answers. It wasn't, it was pattern matching. It was basically looking at what keyw you're using and your pattern matching and that's been around forever. And if you look at AI in general- Airline booking systems, Travel Systems- they've been using AI for a very long time to make kind of make sure, like you know, how do I anticipate Peak demand, how do I anticipate delays and things like that.

If you look at like a credit card companies, like anti fraud detection and so on, they've been using AI for like almost you know, 20 years in many ways where they're like, okay, this person is in locations a, b and c, could they have actually been made this purchase or would they actually, you know, be is like a fraudulent activity and so much of AI has been been happening under the covers for a very long time, like Photo tagging, like Google photos you I could search for, like clouds, and every photo had clouds like, well, that was there like years ago or in you know, iPhones nowadays can like, know, press and select.

Like a picture of somebody. It cops everything else out and makes it C. That's like almost like five, six years ago that was introduced. So we've always had AI. It was just never in an area that intruded upon my areas of expertise. I'm like, yeah, I'm writing code, I'm doing all these things, but a always existed and you know, video games had it when I started seeing, like the Skyrim example. I'm like, wait a minute, a developer basically programmed all of this and now it's able to create this. I'm like, okay, what is it to prevent it from creating a slide deck or a pitch deck?

I'm giving to, like, my leadership teams- how, what is it preventing from writing a document on, like, what is the next big feature we need to do based on this customer feedback we're seeing in all these areas? And that's what got me thinking about AI a long time ago in so many ways, but it wasn't until really, chat GPT came out, like almost two years ago or three years ago, I forget when it came out first. Yeah, yeah, that's when I was like, wait a minute. This is an actual manifestation of the things that I've been ruminating about for quite some time.

We need to solve, you know, get involved in it very, very quickly, and which is why I work on tooling and AI, like you know, at at Google right now it's like, yeah, in the back of my mind always sudden ate away at me, but the speed at which we got, I was surprised. I was blindsighted, to be fair. Like, even though I was thinking about it, I didn't expect it to just happen. And then it happening and evolving and being iterated upon very quickly also completely surprised me.

Sora completely blew me away as well, and I'm only- and, of course, like the music example I mentioned before- like that was a more person, like wow, this actually solved the problem that I actually had, because I had no idea how to make this music Loop work without paying a large amount of money, and I didn't even know, like if I like it or not, whereas here it's like I have infinite choices. I can like change it every single day if I wanted to, and get a new version of it out.

Yeah, and so that's where I'm like, okay, I've hit the limit of my imagination at this point. Like, what can AI do, and what can it not do? I can imagine more things, but nothing is out of the ordinary or out of the realm of possibility at this point. Anything is possible, and so that's why I still say, yeah, I'm blindsided by it. I'm surprised at how quickly things have moved.

1:14:08 - Kirupa (https://www.youtube.com/watch?v=zlAGUJH1-Ek&t=4448s)

Well, I'm surprised that you're surprised, because if you've been thinking about AI for 10 years and you're mind-blown, then I'm 10 times more mind-blown than you are. So the sky is the limit, I imagine.

1:14:23 - Ash Lamb (https://www.youtube.com/watch?v=zlAGUJH1-Ek&t=4463s)

Yeah, it very much is, and that's the exciting part about it as well. If I take a step back and ask what AI is actually going to solve for me, it's a lot of things. One of the things we worked on recently, and I think we shared it publicly, was this idea where you show a cupcake to an AI assistant and it reverse-engineers the ingredients you need. My team built it and we shipped it. Those are really cool, fun examples of things that would never have been possible before. Imagine watching a movie and wondering what vehicle or what food someone is using at the table.

You pause, take a screenshot, send it to the AI, or ask the assistant to explain what's going on in the scene, and it gives you all the details. It could even tell you who the director of photography was for that scene, or what kind of camera they used. Little details like that are now possible because of it.

1:15:20 - Kirupa (https://www.youtube.com/watch?v=zlAGUJH1-Ek&t=4520s)

We'll also have a robot eventually that cooks and creates that cupcake in real life.

1:15:26 - Ash Lamb (https://www.youtube.com/watch?v=zlAGUJH1-Ek&t=4526s)

Exactly. This is the sci-fi version of what people thought the world was going to be like. You remember all those books that said, here's what the year 2000 will look like. Progress up until 2000 was pretty slow, and maybe the next 15 years were pretty slow too, but in the last five years we've gotten exponentially better at realizing the visions of people like H. G. Wells and other futurists from way back in the day. So I'm really excited for some of these things. Let's see where everything leads. It's interesting and it's exciting.

1:16:00 - Kirupa (https://www.youtube.com/watch?v=zlAGUJH1-Ek&t=4560s)

It's out of our control too, so there's no point worrying about it.

1:16:05 - Ash Lamb (https://www.youtube.com/watch?v=zlAGUJH1-Ek&t=4565s)

Exactly. Just embrace the change and figure out what to do with it. Don't try to stop it or fight against it, because that's like trying to stop a giant wave that's coming at you. You might succeed for a brief moment, but you'd be much better off using it as an opportunity, like riding the wave as a surfer in a video.

1:16:23 - Kirupa (https://www.youtube.com/watch?v=zlAGUJH1-Ek&t=4583s)

Oh, that would make for a good visual. I'll write that down. I like that metaphor.

1:16:28 - Ash Lamb (https://www.youtube.com/watch?v=zlAGUJH1-Ek&t=4588s)

That's a good one. Thank you. Happy to. It's been a pleasure talking to you.

1:16:33 - Kirupa (https://www.youtube.com/watch?v=zlAGUJH1-Ek&t=4593s)

Likewise, Ash. Thank you so much. Great fun.


Browse all Interviews with Creative People: https://www.kirupa.com/podcast/index.htm