Ep 81 The Real AI Divide - Enterprise Restrictions vs. Personal Innovation
Watch the YouTube video version above or listen to the podcast below!
Episode Summary
In Episode 81 of Enterprising Minds, the team examines the growing divide between enterprise AI tools and the more flexible systems available to individuals, agencies, and small businesses. The conversation begins with Microsoft Copilot and the frustration created by its licensing structure, inconsistent access to organizational data, and limitations around file creation, automation, and cross-platform workflows.
The hosts then explore the shift from treating AI as a chatbot to using it as an active agent. Instead of merely drafting an email or answering a question, agentic systems can complete repetitive tasks, operate software, update records, organize information, and support recurring workflows. Examples include editing dozens of YouTube descriptions, reformatting spreadsheets, compiling outstanding action items, and turning one information-gathering meeting into several completed business documents.
The discussion also considers the mental adjustment required to work effectively with AI. Users must learn to treat it as software rather than a person, ask it to produce more options than they would reasonably request from a colleague, and decide how much control they are comfortable delegating. That convenience also introduces legitimate questions about security, oversight, trust, and the importance of preserving human judgment.
Finally, the team looks beyond workplace productivity to AI's potential for volunteering, scientific research, historical preservation, creative writing, cooking, and everyday fun. From digitizing historical documents through citizen-science platforms to developing fiction, adapting recipes, and making dry corporate material entertaining, the episode argues that AI's value should not be measured only by how much money or time it saves.
Ep 81: The Real AI Divide - Enterprise Restrictions vs. Personal Innovation Podcast and Video Transcript
Dave Dougherty: Hello, and welcome to the latest episode of Enterprising Minds. The whole crew is here. Yay. Episode 81, guys. We are slowly marching to 100, so by the time we get there, figure out what you want to do for episode 100. Because it'll probably take us a while to, um, to figure out something. But on today's episode I have a particular use case that I would...
I've found interesting, and is related to some of the things that, um, we've talked about previously. And I don't know if you want to start off with this one, Alex, but this one definitely caught my attention. Copilot and why it sucks. Uh, I'm going to let you explain yourself a little more, but that's the hook.
Sure. Yeah.
Alex Pokorny: Yeah. Just a kind of quick thing on Copilot is Copilot's if you are a Microsoft-based company, you know, using different Microsoft products- 365, the rest- you probably, or at least your IT has decided that Copilot is the best thing for you. There are... It has a lot of limitations to it that are built in, mostly around trying to keep things data secure, which is great.
Dave Dougherty: Yeah.
Alex Pokorny: And Microsoft revenue, which sucks, and that's the part that's frustrating to me. The big thing is basically is the, it's how they do their licensing. So there is a free version, um- Mm-hmm ... a Copilot Chat basically, and then there's Copilot. Yes, that's the lovely naming difference, Copilot Chat versus Copilot.
I'll call it Copilot Premium be- more sense if you call it something like that. Right. Um, Premium has a lot more capabilities to it, including file creation. You know, if you want to make a PowerPoint, Premium really is kind of the way to go. If you want to do a, just a chatbot thing, Copilot Chat is just fine.
If your organization has a mix of the two, it can get frustrating real fast.
Dave Dougherty: Right.
Alex Pokorny: Um, the Premium also allows you to create agents. There's a thing called Copilot Studio. In it access to another kind of older system called Power Automate. Familiar with that one. But that allows your own agents, and then you can share it through Teams, through just 365 in general.
Then people, even with the chat version, can use your agents. But it still has a lot of limitations built into it. So the frustrating thing really is, do you want a system that knows all of your data, your chats, your emails, the files, the SharePoint files that everyone created? Right. It can tell you, you know, how's it been going for the last week that you promised.
That is built in with 365, and now you've got this Copilot AI system that gives you access all the, all those, all that data. That's really cool. The limitation is outside systems like Claude, Gemini, ChatGPT for 20 bucks a month can create files PowerPoint files, decks, financial data, you name it, quickly- Right cleanly, and eventually it builds up a memory about you how you do things, your instructions, everything. File types, design, brand guidelines, all that kind of good stuff.
Dave Dougherty: Right.
Alex Pokorny: And then it produces stuff really well. So then you're like do I want to use this system that has limitations to it but has access to all my data, which is cool or do I want to use this outside personal assistant system?"
Dave Dougherty: Right
Alex Pokorny: I think that's a really frustrating model to be stuck in. because for 20 bucks a month you can get something to create files or for more premium license for Copilot, then you can finally file creation. But still it's a little, it's a s- little messy. It still has a lot of limitations to it. Right.
Um, which is frustrating- yeah... because it can run Opus behind the scenes, but it then puts soft environment limitations on top of Opus. That part, I think is the p- part that probably frustrates me the most, is that I know it's running a model that's so much more, but because of pricing
Enterprise Limits vs Chatbots
Dave Dougherty: Constraints- Yep, yeah, I think for me, the Microsoft you can tell is coming from the incumbent enterprise license spot.
Alex Pokorny: Oh, yeah.
Dave Dougherty: You know, and we talked about this in a previous episode where, you know, when they initially launched Copilot, it was Copilot for Excel and Copilot for PowerPoint, and they didn't talk to one another, and that was the biggest miss, I think, because then, you know, everybody was doing AI for my thing.
And you could just see the project managers or the product managers sitting in meetings with each other, you know, and their boss is saying, "Yeah, we have AI for what I'm doing." But it's like, you're right, but you guys are part of an ecosystem that everybody uses, right? That's been the biggest frustration for me because that expectation of, oh, hey, I have Copilot, and I'm seeing it in Teams.
Great. Let me go ask it a question about my calendar, and it's like, "Well, I can't see your calendar." Well, then what are you here for?
Alex Pokorny: That's the data limitations right there.
Dave Dougherty: Yeah. Summarize- and how it's been, yeah ... summarize my emails on this particular topic. "Well, I can't see your emails." What do I need you for then?
Like, stop it. Like, don't even show up. Don't even put the little icon there if, you know, if I don't have it, like, don't even tease me. Like-
Alex Pokorny: Oh, nice ...
Dave Dougherty: Yeah, I think it, it's totally there, and I've been experiencing that a lot in talking to My friends that are on the corporate side and then my friends who are on, like, agency side or, like, individual contractor kind of sides, right?
Like, what people are doing with self-employment, with their own agencies, with, whatever else, there's so much more run room, you know, whereas "Oh, I'm stacking agents on top of each other, and I have my own, like, operating system now." And you're like, "That's amazing. I don't know that I would trust it- yeah,
to the extent that you're doing it, but that's cool that you've been able to, like, really systematize stuff like that. But in order to operate in the same way on the enterprise side, just the amount of, "Okay, I have this amount of context limit per thing. Okay, my master prompt is actually beyond what the context limit is," so I have to then tell it, "Okay, here's the part one of my master prompt.
Here's part two of the entire prompt. Okay, now I'm going to give you a portion of the context I need you to understand in order to do the da, da, da, da, da, da." So now I'm sitting there uploading five different things, and then I hope to God it has enough context window to remember the five things that I've uploaded to then do the task that I need.
You know, and meanwhile, I'm doing that thinking, "If I was just on my personal computer, I could have Cowork do all of this for me like nothing." You know? And there's totally that divide of the people who have started working this way, and then there's the kind of IT departments that are pushing a lot of these things out, where it's like, "Cool, I see what you're doing with your prompt library, but that was two and a half years ago.
Congratulations. We've hit two and a half years ago." I know. And then people who are flatly refusing to even touch it unless it's, "Help me write this email I don't want to write right now." Yeah.
Alex Pokorny: Yeah.
Agents and Automation Leap
Alex Pokorny: The, the difference I think is... Yeah, I think you really hit it there too, especially with the Cowork.
Once it starts to do tasks for you-
Dave Dougherty: Mm-hmm ...
Alex Pokorny: Not just write things in chat, write, you know, the email that you're going to copy and paste yourself, but when it actually opens up Outlook email or Gmail or whatever system, and then sends it on your behalf, that's a whole different era. You know, now you're talking like, you know, multiple users of a laptop or multiple laptops in use or however you're set up- yeah,
but now you're talking like the product change there is phenomenally large. It is; that is a breakthrough once people get to that level. Mm-hmm ... also routine things like having built-in, uh, scheduled events or however you, whatever you use kind of has a different name for it.
You know, every Monday, pull a list of all the different outstanding items that I have from inboxes over the last, or chats together or meeting notes from the last two weeks and give me a list.
Like, you could have that- Mm-hmm ... delivered to you every single day at the start of every day if you wanted. And- Yep ... have it understand when things are checked off and done. Or pull in all the open tickets if you have a ticketing-based system from whatever platform it is, and show me the, you know, urgent ones, up-to-date ones behind, things that I've asked other people that I should follow up on.
All that kind of stuff is completely possible to basically offload- Yep ... which is great to offload. Oh, yeah. Yeah, it is, it is, speaking of, you know, now on enterprise level, the ability to do things and move very quickly.
Alex Pokorny: Hasn't always been such a big deal, but I think with- it is starting to become a bigger and bigger deal instead of, yeah, a small business could totally, you know, switch at the last minute, buy a ticket to sales, and then suddenly be at a trade show that we would spend six planning for.
Yes, absolutely, that is a difference, but we're still both at the trade show. It's still- right... both happen. This though, this is enough of a difference, I think, that it's really eating away at those large kind of companies that are just that slow to change.
Template Hate Use Case
Dave Dougherty: Yeah, I came up with a, a use case recently where it was- you know, like most innovations, it came out of just pure hatred.
And this, I love filling out templates. Hate it. Such a waste of my time. As soon as somebody gives me like, "Hey, here's a 14-page template."
Alex Pokorny: Loathe. I thought you said love.
Dave Dougherty: No. No. Loathe, A-T-H-E. Um-
Alex Pokorny: Yeah, yeah, I gotcha ...
Dave Dougherty: The whole time I'm doing it I'm like, "This is a colossal waste of my time." And you know, it's like the, like when Yosemite Sam is mumbling to himself when he's mad at Bugs Bunny, right?
And essentially that's my workday. And I realized that, okay, I just sat through this meeting talking about process A that we have to do constantly. Then I sat through a training for process B, which is an annual thing, but there's some time pressure because in order for, vendors and third parties we work with, you know, they need to have the information by a certain time for planning.
Okay. Then there's this- Mm ... third process that ends up using the same information that goes against all of these things.
Alex Pokorny: I'm like, "
Dave Dougherty: Why don't we just have a kickoff meeting where we collect all the information for all of these things, and have three outputs from a Claude, from a ChatGPT, from whatever, and then it's just done, then we don't have to do this scramble stuff every year or, every half year, whatever." I don't want to jump when you say, "Hey, fill out this template because it's due in two weeks." Nah, I got my own thing going. I don't need to.
Alex Pokorny: Yeah. Unnecessary little
Dave Dougherty: tasks. Yeah. And but it was interesting because it was, it hit me, you know, that, hey, okay, these are all really important things, right? I mean, these are business-critical activities. However, they're taking up three different departments’ worth of time to do these things. That, now that I know what's possible with a GPT, with an Antigravity, with, you know, a Claude, it seems like such a colossal waste, you know?
And it's one of those, like, I don't know about you, but when we talked about, you know, going from treating it as a chatbot to then going into the agentic side of things, I don't know about you, but in my experience, it's taken me a while to make that leap and not just treat the agent as if it were a more capable chatbot, right?
But to actually sit and go, "Okay, I should actually map out this project." And then you have to add different sections to make the agent work better than just the chat-
Dave Dougherty: Cause you're less involved. Did you experience that? Like, was there some cognitive dissonance where you're just like, "Man, I don't know about this agent stuff"?
Or were you just straight into it because you're brilliant?
Alex Pokorny: No. It was... It took a minute, definitely.
Thinking Like Software
Alex Pokorny: There was a really good interview of Ethan Mollick a while ago, and I was doing this, maybe it's the Minnesota nice thing. But his conversation helped me kind of break through it, was instead of asking for, you know, give me two examples or one example or three, ask for 15.
Right. I need a new, uh, subject line for this email. Give me 20 ideas.
Dave Dougherty: Right.
Alex Pokorny: That would be something I would never ask of a human being, because that seems like an inordinate waste of their time or just- Right ... an unreasonable ask. But this is a piece of software, and that's the thing. It's like the more that you can think of it as this is yet another application that's on my laptop, this is yet another application on my phone.
It's- Right ... just an app. And I think any time we start to personalize it too much... my wife caught me once when I said Claude as a he, because I always say it.
I don't he. She caught me once. "Well, you just said he with that." I was like, "What?" I was like, "No." Like, "Are you kidding me?" Like, I would never...\
Because that's the thing is like once you start seeing a piece of software as an individual, which, I mean, if you comment on S% Reddit, there's some people who absolutely love bots. I mean, they're- Mm-hmm ... they're best friends. Mm-hmm ... then there's a mental shift there. And if you just see it as software that's just running, like I run Crawler, like today I wanted to access and test 36 URLs, and it's going to run.
I mean, it's just that's... Would I ever manually test 36,000 URLs? Goodness no. But it's- That would be your
Dave Dougherty: entire year.
Alex Pokorny: No, absolutely. Yeah, exactly. I'd spend my next week opening tabs. No. Just to see URL patterns that may or may not fail and exist. No. No, no, no. Not going to do that. So that mental break helped. The sticking point for me when I, with the agentic side was basically is on control, which I even asked, like, "Am I giving you too much control?" And it suggested ways to break, and then it just broke systems by breaking it. That, that was a little strange, especially when you see the thing pop, you know, like your Gmail calendar, and then three seconds later it pops up in a new event and it starts- Mm-hmm
you know, working on it. Like- Any kind of... Whenever I've run software in the past that does any kind of computer control is just, it's creepy watching it just- Right ... move your mouse around, and suddenly there's this, like, ghost moving your mouse and clicking things and stuff, and it's like, it still has to use the mouse, and it still has to click things, so you still get that, you get to watch it.
... That part, I definitely, I had concerns about. I wasn't like a lot of people with, like, OpenClau- "Yeah, here's my bank account." Yeah. Uh, "Come up- So- ... come up with an investment plan and invest my money, and here you go. It's an AI." I didn't do that. I... It has a lot of security concerns, and those are enough to kind of, to stop, but-
Dave Dougherty: Yeah.
Do you remember- I've given it access to some things ... when the internet first happened, and everybody was scared to put their credit cards online?
Alex Pokorny: Yeah.
Dave Dougherty: You know? But I, yeah, I still remember that, where it was just like, "No, somebody's going to steal it." And guess what? They have, multiple times over
Alex Pokorny: The- yeah.
The encryption's gotten a lot better over time, and that was, that was kind of the, I think the lesson from it is just, like, you gotta hold off with some of this to understand-
Alex Pokorny: The different angles where it can be positive and negative, and also judge, for yourself- right... of where's that level of, like, security that you need to feel.
And I know people that still have, you know, concerns about that some credit cards are built, especially like, stripes versus the chips and stuff like that.
Dave Dougherty: Right.
Alex Pokorny: Some of the ways that- right... you can spoof those things. Which valid to some degree. So you have to kind of figure out, like, you know, what's your level of convenience you're willing to give up,
Dave Dougherty: right.
Alex Pokorny: Security versus convenience.
Dave Dougherty: I know for me the, to your point the seeing the computer just do stuff.
Gemini Browser Automation
Dave Dougherty: I- Yeah ... having, my experiments with that have been largely through the Gemini via Chrome, that little browser extension piece, and it's specifically for individual tasks. So perfect example. Mm-hmm.
Like, when we launched enterprisingminds.com. I had to change all of the Bitly links in all of the episode descriptions for how- That's a lot ... I think it was, like, 69 episodes at the time or something like, So, I mean, it was a lot because there's an audio version- Yeah ... a video version, and yada yada.
Alex Pokorny: Sure.
Dave Dougherty: So I just, I went ahead and I'm like, "All right, let me try this Gemini for Chrome thing." And I said, "Hey, here's my YouTube channel. Here's the text that you should be looking for. Here's what you should change it to. Go." And then I jumped on a meeting, for the day job and just sort of, you know, out of the corner of my eye looked over and it was, pop, pop, pop, pop.
And it went through all 69 or whatever. Nice. Uh, it took about four hours and I had to tell it to do, "Okay, next batch, next batch." Like, it had its own limitations, which I was kind of nice to see. Like, yeah, "Hey, we've reached the limit of the number of actions I'm allowed to do." Okay, that's good to know.
So I actually, like, kept- Yeah ... track of it to say, okay, it actually took like four hours to do this, but honestly, it probably would've taken me that amount of time to go line by line by line. So it, you know, it wasn't, like, unbelievably quick, but the fact that I could just set it up and let it go, and then come back and say, "Yep, that's done," that has been great,
Alex Pokorny: yeah.
Dave Dougherty: Or a good use of it ... even, like, the spreadsheet formatting, right? Like, hey, uh, this spreadsheet's set up for me to track stuff, not for, Airtable to talk to YouTube, to talk to, you know, all these other things. Format it so that robots will actually understand what it is, and, like, it just does it all.
And that, that's been great too because what... Again, what a colossal waste of my time to reformat a sheet with data that already has data in it, but I mean, that was- Yeah ... a large part of work for a long time.
Alex Pokorny: Like- Yeah, and you're seeing that change.
Dave Dougherty: So-
AI for Volunteering
Dave Dougherty: You have this other interesting idea that we've talked about a little bit prior.
But AI for volunteering. What
Alex Pokorny: are you- I have a practical example.
Dave Dougherty: Fantastic. What are you thinking with that? Help me understand that.
Alex Pokorny: So, yeah, I've been on this, have been for a long time, of this tools, these tools are so cool- Mm-hmm
what is the best way to use them to kind of help humanity? And the tools sometimes a little struggle a little bit, especially if you've got a long history with them of getting outside of your own world, but also greater global world. Like how do I... How are these things going to be useful? Like, can I get this discover cures for new diseases to tell it to or, can it do?
Um-
Dave Dougherty: Right ...
Alex Pokorny: At Gemini, ChatGPT, and Claude, I tried this at least twice in the past where all of them kind of did like the AI council thing where you're in a prompt for all of them and see what, and kind of critique them, have them- right... critique each other's outputs and stuff.
Dave Dougherty: Right.
Alex Pokorny: It, there's limitations there, but the systems are getting a lot better.
Citizen Science Digitizing
Alex Pokorny: So I want to, again, and I was thinking about I do some volunteer work through the thing called Zooniverse. So think of like Universe except Z-O-O at the beginning of it. And what it is, is citizen scientist programs. So NASA has a ton of them. ... NOAA has a ton of them, which is the national, what, organization- Oceanic
air and water. Yeah. Yeah, exactly. Or you're right, it's something else, but it's weather-based kind of data. So they're asking people from certain areas to take photos of clouds or rainfall, or with the NASA projects, it's a lot of telemetry and telescope kind of data that's coming back. There's tons of it.
Right. And while they can use some basic machine learning to delete some stuff, there's still a ton of stuff out there that still has to be labeled and identified. Might watch a short GIF and you swipe left if there's nothing, swipe right if you think there's a, which might be- Hmm ... a brand new planet, and you can write your name on it, which is pretty cool as a measure- That is cool not discover, but public- Right ... thing out there. And, and there's been papers written with all these citizen volunteers being listed in the papers too, so it's a really cool, like literally on your phone, swipe left or swipe right, you could be in a scientific paper someday. So- Yeah, so I came across it again kind of through a different way.
But I kept thinking, I was like, "Okay, how..." It's a little difficult because it's a GIF that you're swiping left and right. There is a desktop version that actually you have even more tasks, more complex tasks that it gives you. A little difference between the two systems. But how do I get Claude to do this?
Like, there's CoWork now. Last time I ran these tests, CoWork wasn't a thing.
Dave Dougherty: Right.
Alex Pokorny: Um, but what I got into was transcribing and digitization of files. So last night, I helped digitize the JFK assassination report, the Warren Report. So I digitized Jack Ruby's who was the guy who killed Lee Harvey Oswald, who was the assassin.
Um- Right ... his bar receipts, the stuff that they found on him. I typed up all that stuff using screenshots that I then sent to Claude, which then I verified and then used as the transcript and was able to archive those random documents and things. Uh, he got a parking citation, which I wrote up I also did an 1880s, uh, ship manifest that was, uh, one that hadn't updated yet.
There's a lot of, uh, World War II and Royal Navy vessel, uh, reports. Hmm. How many people there, what rank, their names. There's a lot of people that are also kind of named in things because there has been no digitization of their records, so-
Dave Dougherty: Right ...
Alex Pokorny: records. There's a ton. Revolutionary War records, a ton.
There's a bunch of nurses reports that say people's name ever known to be in the war who died. Somebody's historic potential grave things like that. So a ton of programs out there- So it's your fault that I've- ... for transcribing ...
Dave Dougherty: been getting pinged by Ancestry for, new information on my triple great-grandfather.
Alex Pokorny: Found we know something.
Turns out he lived to a ripe old age, and he was a bastard.
Dave Dougherty: You know, there's a surprising a lot of those in my family.
That's really cool.
Alex Pokorny: Yeah. That's
Dave Dougherty: Interesting.
Alex Pokorny: It was cool, and there's, there's other programs as well. A lot of the citizen scientists kind of is... There was some sort of ML sort this stuff too. There was some kind of like machine learning system that was run on, but it was an outdated, not great one.
Uh, National Archives used to use one from Amazon, but it's a pretty old, at this point, system that missed a lot, and it had- it struggled with like 1880s cursive. Um- Interesting. Yeah ... which new systems can totally handle. So yeah, it was fun. It was literally in half an hour I started transcribing old documents to help with the digitization of the assassination report.
That was big deal, the Warren Report, adds a whole name to it, people.
Dave Dougherty: Right.
Alex Pokorny: And there's pieces no one's ever typed up Yeah, so fun.
Dave Dougherty: Now you're going to be in the whole conspiracy world by actually handling the- ... his ticket.
Alex Pokorny: No. It, it was kind of cool though, just because, the manifest or in the World War II stuff-
It's of... Like, some of these also are diary entries from people from World War I- Right ... or the Civil War, and you're typing up their diary notes. Actually just kind of cool reading to begin with to offer something in return to it. Right. Kind of a cool experience. So I think once we see these AI tools also become more available and also, like, ability just to become more known, not just the techy elite I would s- kind of set aside, but for everybody- Right
else as well to be using these tools on a frequent basis and just think of, you know, Microsoft Word as, "Yeah, I can do spell check. This I can also use to transcribe." Mm-hmm. Once it becomes, like, that kind of mentality, I think it's going to be really cool because there's a lot of philanthropic abilities with these tools that could be helpful.
So yeah, it was fun. It was a fun little random evening.
Dave Dougherty: Yeah.
AI for Fun Creativity
Dave Dougherty: It's an int- it's an interesting play too, because, you know, so much of what we see about AI is just, you know, be more productive, make more money.
Alex Pokorny: I know, and
Dave Dougherty: that's- That's just
Alex Pokorny: part of it ... you know, I... Yeah. You know, like my fitness or even the book writing thing, which is, it's got a little bit more to it, but other random, my plant watering schedule, I have so hard on Reddit and all these other places for things like, AI in personal life ideas- concepts. People just putting down kind of the random stuff that they've tried and stuff like that. There is next to nothing out there that I can find. So many people are so hardcore into creating campaigns- Yeah ... um, blog posts, you name it, that like the AI for fun, like-
Dave Dougherty: Right ...
Alex Pokorny: there's a whole lot of cool you can do with this stuff.
Like-
Dave Dougherty: You know, it's- ...
Alex Pokorny: it doesn't seem to have hit that level yet.
Dave Dougherty: It's hilarious. Like, you and I have talked about, you know, creative writing, but,
Alex Pokorny: Yeah ...
Dave Dougherty: I kid you not, since AI has gotten to where it is this year, I have four friends who are working on novels. Four. And the funny thing is I was the only one with a creative writing degree.
So I'm like, "All right. Hey, you f- I gotta get in on this."
Alex Pokorny: What are you hacks doing?
You know- You know- ... I started taking a class, uh, online class recently, uh, writing fiction books. Um-
Dave Dougherty: Mm-hmm ...
Alex Pokorny: and it, it did really good because I started critiquing my own writing that I've been doing with Claude as an editor.
Dave Dougherty: Right.
Alex Pokorny: And realizing, some of the things that the professors were saying on the, in the class were like, "Yeah, this is a sign of an amateur."
I'm like, "Oh, yeah. Yeah, that's like my entire book." Mm. "Yeah. Yeah, yeah. I'm there." Which was funny too because Claude didn't know it. I've asked Claude m- a number of times, like, you know, judge my writing against-Professionals. Like, you know, all these other kind of things. And there is... The funny thing is AI really trained on really crappy bestsellers novels.
That's a huge part, still data set.
Dave Dougherty: Mm-hmm.
Alex Pokorny: So its bar I think is really, really low.
Dave Dougherty: Right.
Alex Pokorny: So-
Dave Dougherty: Right ...
Alex Pokorny: it wasn't real helpful, I realized, for some of these el- elements of, like, fiction writing, which, you know, setting the scene better, descriptive, which I wasn't. You know, things like that.
Friction and Skill Building
Dave Dougherty: Well, and that's, that's part of the, a recurring theme that we can explore in the next episode, which is the need for friction.
I think I'm hearing more and more, and more conversation around, like, if you don't go through the process yourself, you're not going to be- yeah... good enough to know what the output is. So, like, what do you want to be good at?
Alex Pokorny: That's
Dave Dougherty: good. Yeah. And then leverage- 100%, yeah ... you know, leverage the AI for that, right?
So in my case with the creative writing, I had written over 20,000 words for my undergraduate capstone. So it was about eight chapters of- Sure ... something. And I started reading through it, and I'm just like, "Whoa." Now, granted, that was years ago. But the interesting thing was leveraging b- all the different AIs, to your point, using, like, the council idea of saying- Yeah
"Hey, I've had this idea of how, like, James Patterson does his books." Sure. And he does, like, 70 or 80-page outlines of, like, "This chapter needs to do this. These are the characters involved. It moves the story forward in this way." Wow. And so da, da, da, da, da, da, da. Okay. There's a very specific name for what it's called.
I will give it to you after we are done recording later. But I essent- I uploaded that,
Alex Pokorny: Yeah ...
Dave Dougherty: Outline idea. And then I uploaded the or- the original writing, and then I said, "Create an outline in this style for me so that I can see the beats and what the main themes are and whatever else." Yeah, yeah.
And then I started having conversations around the characters, and then like, you know, I was like, "Okay, this is cool. I like this direction. I don't like that, but it doesn't seem like anything's at stake. What's, what is this book even about?" because it's, it's not like, you know, Godzilla's not coming and destroying the city.
There's not a, you know, super secret spy. There's, you know- Sure ... nothing, like- Sure ... immensely at stake in it. So, so what is it? And so it took a while to go back and forth on, this book is about X, and each character does this in service of that, and the conflict is actually the fact that they're operating under their own ideas of what's correct, and that's what screws it up for everybody.
Right? Hmm. And so mapping that out, and then to your point, I was an early writer, so it's very dialogue heavy. So then the other day I sat outside- Yeah ... because it was a super nice day while, you know, the neighborhood boys were playing basketball on my driveway, and I just started writing scene work for, this is the town, this is the type of town, these are things that make- Sure
it interesting. Right? You know, and just- Yeah ... been running through it. Is it done? Is it perfect? Is it whatever? No. It's a page and a half about a town. It's just to get the ideas out and fill it, and feel it out. Yeah. And it's handwritten, right? Because I want that- Yeah ... I want that friction to actually play with the idea and build it out, and do that.
But it's been really interesting to talk to, you know, you and my other friends about writing and, and I honestly think it is only possible because we now have time to think or explore on those ideas as opposed to what we had before. Yeah.
Closing and Sign Off
Dave Dougherty: So Ruthi, I know you had to go deal with some stuff, like some urgent things, so we didn't really hear your voice today.
Do you want to do any comments in the next minute and a half other than saying hello?
Ruthi Corcoran: Hi. Thanks for having the conversation in my stead today. It sounded like it was a real good one, so I'm going to go back and listen. Um, the only thing I will add to this on Alex's point about, you know, using it for fun or personal things, um, is just that, uh, last week I had Claude help me expand a Mary Berry recipe for a- Mm
pesto pasta. I said I wanted to add chicken to it. Please create the recipe in a way that Mary Berry would create the recipe. Mm. And it did a lovely job.
Dave Dougherty: Oh, that's cool.
Ruthi Corcoran: Mm-hmm.
Dave Dougherty: Nice.
Ruthi Corcoran: Mm-hmm.
Dave Dougherty: Oh, speaking of fun things, I think I sent this to you via text, but it is hands down my favorite thing to do now for a really dry corporate reading.
Tell w- the AI- ... of your choice to take whatever thing it is, but do it in the style of a Little John song. Oh my God, it makes financial documents way better.
Ruthi Corcoran: I am so glad you shared that because- ... like, some days you just need a little, little pep in your step, and y- you gotta get through emails, and that's a much better way to do it.
Dave Dougherty: Uh, yeah, I love in the, the one instance that we had, I love the fact that it was like, "We got corporate compliance in the house." And they never get a shout. That team never gets a shout. No.
Alex Pokorny: They're excited at least.
Ruthi Corcoran: Oh, that's
Alex Pokorny: true. Angry shouting, never happy shouting.
Dave Dougherty: Yeah. So on that note, go have fun.
Go do something silly with it. See what happens. And, uh, you know, for anybody in the States, have a happy Fourth. Hopefully you had some time off, uh, with family and friends and got away from stuff for a while. With that, take care. Like, share, subscribe. It helps. Enterprising Minds, Pathways, all that stuff, all that good stuff.
And we'll see you in the next episode. Take care. Cheers.
