Ever imagined a world where AI liberates us from mundane tasks, freeing us to channel our unique skills and creativity?
Galen Low is joined by Kelli Korducki—Independent Journalist (wrote for The Atlantic, The New York Times and NYT Magazine, Insider, The Guardian, and more)—to talk about the use of AI tools and large language models in digital industries.
Ever imagined a world where AI liberates us from mundane tasks, freeing us to channel our unique skills and creativity?
Galen Low is joined by Kelli Korducki—Independent Journalist (wrote for The Atlantic, The New York Times and NYT Magazine, Insider, The Guardian, and more)—to talk about the use of AI tools and large language models in digital industries.
Hey folks, thanks for tuning in. My name is Galen Low with the Digital Project Manager. We are a community of digital professionals on a mission to help each other get skilled, get confident and get connected so that we can amplify the value of project management in a digital world. If you want to hear more about that, head on over to thedigitalprojectmanager.com. Okay, today we're doing something a little bit different. Today we are getting a journalistic perspective on the use of AI tools and large language models in digital industries. So things like what these tools are good for today versus what they're not quite good at yet, and what skills digital professionals need to focus on to stay ahead of the AI curve. Joining me today is Kelli Korducki, a seasoned journalist who has written for a whole slew of publications that you probably have on your desk right now and who has recently been deep diving into the use of large language models in the tech industry. Kelli, thanks for joining me today.
Kelli Korducki:Thanks for having me.
Galen Low:I love your background. For listeners who are not watching the video, Kelli's got a lovely plant collection, bookshelf, and very nicely lit New York apartment. Pleasure having you on the show. I mentioned at the top, this is a bit different. I don't normally have journalists on the show, but I thought this one was really important because there's so much going on with AI and large language models that spreads across different industries, across the digital sector, across tech industries. And I really wanted your perspective because you've been looking across and aren't just in the sort of like digital project management sort of blinders that I am. And I thought it could be a really good conversation. But I just wanted to give our listeners some bearings. So, I mean, you have an incredible resume, at least I think it's an incredible resume as a journalist. You've written for publications like Cosmopolitan, The Atlantic, The New York Times, Insider, The Guardian, and that's just to name a few. You've also written a book on the history of breaking out, which I am very keen to read. I haven't read it yet, but I'm very interested in that. But what struck me when I was just looking you up was that your range is pretty immense. And lately, I know that you've been diving a bit into AI and specifically just like how it can be embraced to make the future of work just a little bit more human. So I thought I'd start by asking maybe the obvious question, just like what pulled you into AI? As a journalist, I understand that maybe it's just a wave you have to follow, but when you and I chatted earlier, I could really tell that there's this passion about it and it's it resonates with you. So how did you get into it and what pulled you in?
Kelli Korducki:Sure, that's a great question. Well, yeah, I think I got into AI through natural curiosity, honestly, self interestedly, not because oh, it's this is new, exciting thing that people need to cover, but oh, is this going to take over my career? Am I going to be completely non-viable in 5, 10 years? Should I retrain? And so I really, when the kind of generative AI wave really started taking off, which would have been a little over a year ago, this was like fall of 2022. So it's wild to think how much has happened so quickly. But that was when I started really paying attention to it. And part of it was just I was working as an editor for the Atlantic, editing newsletters. And one of the writers that I was editing is Charlie Warzel, who is a tech columnist. He's a now tech staff writer for the Atlantic. At that time, he was writing a technology newsletter. And he's previously been at the New York Times and BuzzFeed writing about tech. So, he was the first person who, I think my first week on the job in August of 22, he was writing about DALL-E and AI art. And I was just like, oh, this is really interesting this opens up a lot of really interesting questions about what the value is of the human touch. What do these AI tools bring to the table? How does this change the way we think about even like human knowledge and human intelligence and the value of human beings? So then fast forward to GPT-4 gets released in March, mid March of this year, 2023. And yeah, you immediately hear all these incredible things like it can pass the bar exam and it can score really high on the SATs, like standardized tests, that it can write an entire term paper on command if you just give it some instructions. So all these things, of course, piqued mine and everybody else's interest. And immediately, I became a paid GPT-4 subscriber right away, as soon as I could. Started playing around with it, and also just started trying to understand how it works. What actually, I don't have a computer science background. I don't have an artificial intelligence background. I don't have a computational linguistics background. I was an English major in university. So I really did have to learn all this stuff in a very basic layperson's way. And the more I learned and the more experts that I spoke with for stories, and I would just like, even not for stories or just curiosity, the consensus really seemed to be that there are simply some things that a machine cannot do, an AI cannot do. AI cannot react to people, an AI cannot respond to a new situation or a, an unpredictable circumstance, and then make a judgment call. They are derivative models. Large language models operate probabilistically. So, you give them a cue, and then they'll give you the response that has the highest probability of seeming like the right thing. And in a sense, that's very cool because, that is how human deductive reasoning works too, to an extent, but there's limits to that. It can't check itself. It can't follow a train of thought. It can't come up with something new. And to me, that's like the most valuable part of what I do. And that's actually, the thing that's taken the longest to develop. I was always a good writer at the sentence level, but like becoming a good thinker was something that I had to develop over time. And yeah, AI can't think. It's clever, it's not smart. And there's a difference between the two things.
Galen Low:And I like how your perspective has centered around work and around the value of human knowledge. And I think even just in any profession I suppose but definitely in project management and I would suspect in journalism as well. It's easy to reduce it down to be like, Oh, you put words on a page? So does this large language model. But actually, there's something else going on there that is the smart bit, beyond the clever bit. And I think everyone's job probably has that. Actually, when we were chatting before, I got this distinct sense that, you know, we are talking about AI, but specifically, most of us are talking about large language models right now. That sort of probabilistic, fill in the blanks based on what's been done before. And from that perspective I agree it is not generating new things. It's generating stuff, but not necessarily new things. And then, I mean, the sort of dot dot dot beyond that is there's probably heaps going on in AI that is getting pretty smart. I was listening to what was it, Diary of a CEO, there was a someone from Google whose name I'm blanking on right now, but we're talking about just how quickly AI could get smart. But, on this day here in, November 2023, I don't usually timestamp these things, but anyways, it is a moment in time where we're like, okay, what we're dealing with right now is like these tools that are using large language models, and it's changing the way we think about ourselves. And it's existentially scary, I guess? But I like that whole thought process of where you landed, which is okay, but I'm smart, the large language model that I use, I was curious about, and I need to embrace it. But, I need to treat it as the clever assistant more than I need to treat it as something that's going to make me obsolete, I guess, which I think is interesting. I mean, it's a funny thing because, in all the conversations I'm having these days the big question is okay, well, how long before AI takes my job? But even though that is the question that we all think of, and I think even, in your journey, it occurred to you as well, but is that even the right way to be looking at it from? And if it's not, then what is the most productive way to look at this and answer that question so that you can move on with it?
Kelli Korducki:Yeah. So the quote that I always go back to when I hear that question, is AI going to take my job, is this AI expert that I spoke with for a story that I wrote for The Atlantic earlier in the fall. Told me, AI does not replace jobs, it replaces tasks. So, if you think of a job, anybody's job is made up of all sorts of tasks. Some of them are, grunt work, busy work, administrative stuff, like sending follow up emails, putting stuff on a Google calendar, managing a budget spreadsheet, like these kinds of things that may fall into somebody's job and to many people's jobs, but aren't the main thing of the job. Like the main component of the job is in the case of project management, it's like managing people, making sure that people are working together in a way that makes sense, that is achieving a target within a certain time frame and ideally within a certain budget. And AI can assist with certain tasks that are associated with that, like maybe scheduling or figuring out some of the budget details. But at the end of the day, this is a people job. It's about people being able to be in a room or in a virtual environment and to look at the other people in a team and to be able to say, okay, like this person works well with this person. This person needs to be doing this kind of thing. This other person is really strong in this capacity and to Tetris together all of those different parts in a way that makes sense to achieve a goal. And that's just simply not something that any language model is currently capable of doing or may ever become capable of doing. And that's also the most important part of a project manager's job is that people part. That is what is the art of project management. So yeah, if you're thinking about AI in terms of project management, you really can think of it as like that handy assistant that kind of takes over some of those, I wouldn't say lower skill tasks, but less strategic, less person oriented tasks and makes sure that they all play nicely together.
Galen Low:It's funny because I mean, I think in some ways, maybe we all think it in our heads, and I'm sure it happens in some conversations, but like the loudest voices and the most sort of public conversations. I don't hear a lot of, hooray we get to just relax, and not do scheduling, and not figuring out budget variants, and not do these tasks that most people didn't even want to do in the first place. But now take pride in it, and now we take so much pride in our work, our professional lives, and nothing wrong with that. But so much so that when this technology comes along to be like, you can do less, you don't have to do as much. We're like, Oh gosh, I'm afraid. It's a very sort of interesting mindset of the way we are protective about our roles and our professions and our careers. It's just been interesting. I don't know. Have you had a lot of conversations where folks are like, No, this is great. Please take over those tasks because I can do other things.
Kelli Korducki:Oh, yeah, that's actually a really good point that I feel like we don't really hear. But it's true. My experience, personally, and also with everybody that I talk to, and most of my friends work in journalism or some kind of journalism adjacent, like for almost all of my friends are creative professionals. And so, yeah, the funny thing is, instead of Oh, I don't want this machine to mess with the things I can do. Everybody that I know is Heck yeah! Please take this thing that I don't need to do. You want to transcribe my recording? Amazing! Please go ahead and transcribe. I think people really at least the people that I know and myself I think people tend to recognize it for what it is, which is it's not going to be able to do your work for you in terms of the work of actually making a new thing that didn't exist in the world before. But it's very useful for doing kind of, yeah, mundane, not even administrative, what's the word, like more rote tasks I started using an AI transcription audio transcriber almost immediately. As soon as that became available, I began using AI transcription for my audio recordings for the interviews that I was doing because I'm recording in a given week, three, four, five hours worth of audio. And if I were manually transcribing all of that, that would be my whole life. I would not have time to actually write. And actually writing is what I'm being paid for. I'm not being paid to transcribe interviews. So, I think that a lot of us, at least everybody that I know sees it like that. It's really, really useful for summarizing really, really long, if I'm, if I have to summarize like a few research papers, like scientific, published in scientific journals, that's, a 40 page document times whatever number that I have to go through. It's very useful for me to feed it to a chatbot and ask it to give me the main points in layperson's terms. And, obviously oh, at least I personally best practice for everyone, you should definitely double check because these things are not a hundred percent reliable and they do make things up. They hallucinate. So you have to fact check. But yeah, it's really helpful. It's a huge time saver and it frees up the time to do the work that is actually the work. That is the work that I am uniquely qualified to do.
Galen Low:I like the framing on, well, the people in your circles are creative professionals, but I think therein lies the key, right? Which is that a lot of our, these professions, they have a creative aspect to it. Like you were talking about things that are rote versus things that are net new, right? And I think even project managers, I think a lot of us actually do think that we are being paid to do some of these tasks that happen in a spreadsheet or, just like writing status updates. But I think, for the folks who do, they're missing the point. I think the point, like your point earlier, I think is very well taken in my books, which is that, yeah, the sort of at the center, the core of project management is about people. It's about, tackling new challenges with people every day, things that haven't been done before necessarily, and then, problem solving yourself to a successful delivery of a project, which in a way you could say is all creative. That's all the creative stuff. That's the stuff that, a project manager is uniquely qualified for. So I think that resonates with me, as a sort of like the creative aspect of any job. I think it's the thing that should get amplified through this, we would hope. I mean, that does lead me to, the next question, you've been talking to folks all across tech, developers, folks all across the tech industry, but in terms of just yeah, like the value of human knowledge, the creative side, the smart bit, like what should project managers working in digital be doing right now? How can someone use this today to their advantage? What should they be focusing on?
Kelli Korducki:Well, I think the first and most maybe obvious answer to that would be just to start playing with the tools, start making friends with the tools, incorporating them into your work life. I spoke with this really interesting project manager who's just like a veteran. I think she's been working in the field for 30 plus years. She's based in Kelowna, British Columbia on the west coast of Canada. And she was telling me that she thinks that it's the best thing to happen to the fields. And this is somebody who's been working in the field for like forever. And she's just finally, we're going to have people actually doing their homework before they go into a project and actually maybe plan. She thinks that it's going to free people up to basically do their homework about the different moving parts that they need to work with. And that it'll improve project success rates. So, yeah, I think that maybe an approach or a way to think about approaching it is, like, a buddy for doing the creative strategic stuff in a more linear organized, less scary, more successful way.
Galen Low:Yeah, no, I really like what you're saying just in terms of making friends with the tools I think is huge. And then, if I'm picking at what you're putting down this is homework, right? You mentioned homework. Homework is in skilling up learning about some of the new technology in place that should be able to help you do your job. But I think the other thing I picked up there was like, the homework of things you probably should have been doing as a professional, as a project manager. But maybe weren't able to because, you're always bogged down with these tasks and now you need to rethink what your value is and the sort of preparation and the homework you need to do to make a project go well. And, I would suspect that it's probably not, spending an hour every week, putting together a status report, but maybe that whole aspect of the job that was like, and then literal homework, right? To go and be like, okay, well, what is this project about? What are the goals about what value am I bringing? And what value is the team delivering? What business outcome is this driving, how does this fit into the bigger picture rather than the is this milestone the right date? Are we going to hit it? That's important too, but maybe the real sort of successful individual at this moment in time is going to be the person who is being creative about, oh, okay, what can I do now that's different? If I've made friends with this tool, and it can help me with some of my tasks. And, I do think that a lot of people are thinking that. But it is as someone who is creative yourself creativity is hard, right? Especially when you don't think of yourself as a creative person, you've been doing things, quite repetitively, project over project, you're I've got my method and then you need to change it. And actually that is the hard part is coming up with a new thing and not really having any foundation already of what this looks like.
Kelli Korducki:Yeah, definitely. But to that, I would say, any knowledge work, which I don't think anybody would disagree that project management is knowledge work. Any knowledge work is creative work because at the end of the day, it's about translating concepts into executing something new. And anything that's remotely innovative or advances any kind of goal, the achievement of a target is going to be creative because at the end of the day, problem solving and strategizing a path to get something done is creative work. Creativity or like the, to be an artist is, people often talk about their creative practice, their writing practice, or the painting practice. Because creatives, like artists, tend to think about the goal of kind of the process of making as the point. It's about the whole thing that happens from beginning to end. Anyway, that's a long winded, widening way of saying that I think it's pretty actually analogous to project management or any other kind of knowledge work. We tend to think about being creative as some kind of special sauce or magical process, but at the end of the day, it's actually just about doing the practice of asking questions and trying to figure out different ways of getting toward where you're going. I mean, I think that probably I would imagine that project managers are already toying with their processes from project to project. I would imagine that no two projects are the same and that part of the work is fiddling with that process, tweaking it and making little adjustments in order to facilitate the execution.
Galen Low:A hundred percent. This may be a little bit of a left turn, but you know, I'm thinking through this notion that maybe not every project manager thinks of themselves as creative. And I know that, the process of making something, as a creative person, it's a bit of a different approach. So I just wonder what are some of the things that you do to supercharge your creativity or get yourself inspired or really generate a new sort of way to approach whatever you're in the process of working towards or building. What are some of the rituals that you have to put you in a creative mindset that maybe a project manager could adopt?
Kelli Korducki:Well, this is gonna be, , the most boring answer, but if I'm really stuck, I'm usually gonna try to take a break. And then honestly, if I'm reading something that has nothing to do with what I have to work on, I'm gonna return to my assignment or my project a million times refreshed. The quality of my thinking is gonna be better, the quality of my actual writing is gonna be better. It's just sometimes you just need to step back and put your mind on something else that is still active thinking something that engages thought, but isn't work.
Galen Low:That's fair. Yeah, actually, I was thinking, that's, it's an interesting approach too, because I was like, I was also thinking about inspo, right? Do you, would you pick up actually something similar and related to inspire you? Or does that kind of throw you off actually because it puts you in a comparative mode?
Kelli Korducki:Definitely. I mean, like sometimes I would call that research.
Galen Low:Yeah, fair. Yeah. Okay. There you go.
Kelli Korducki:But yeah, definitely. To get more information. Because sometimes being stuck is the result of not having enough information, for sure. So getting more information is always helpful.
Galen Low:And honestly, like project managers can be bad at that. We're like, in our silo thinking we have to figure out every problem by ourselves. And there isn't necessarily a bookshelf of, other people's projects that we can leaf through to get some ideas and, do our research. I mean, we certainly do in our community, right? These are conversations we have, day in, day out. That's a neat way to look at it as research. I hadn't really thought of that. But, I don't know, I was just thinking I'm on this thread of yeah, how can project managers, adopt more of a creative process and think of themselves more as creative professionals because it will help work in tandem with some of the tools that are based on large language models that can take on some of the tasks that are, arguably less creative. But then they need to be creative about thinking about what they ought to be doing instead now that these tasks are being taken over, or at least, have been delegatable to a large language model. It's an interesting thought.
Kelli Korducki:Yeah, well, it's interesting. So the woman that I mentioned earlier, who the project manager based in Kelowna, British Columbia. She was just describing how she's using like this, I forget what it's called already, but it's a whiteboarding tool, I forget which one. It may be Mural was the whiteboarding one and then there's another one that's like really good at Gantt charts.
Galen Low:That would be Tom's. Yeah.
Kelli Korducki:That's Tom's. Okay, so Tom's was the Gantt charts, Mural was the whiteboarder. But anyway, the way that Annie McLeod was explaining the process of using a large language model in a project based setting requires, well, you have to give it prompts, you have to give it cues, right? So, you can't just it's not like it's going to magically generate a thing. You have to think through the process as you're feeding it the instructions. So, it ends up being this brainstorming tool. It's like a collaborative brainstorming tool. You have to think through what you're doing and what you're going for, then adjust as you go along. And it ends up making for just naturally a creative iterative process. These tools might actually, in some respects, help individual project managers be more creative in the way that they approach their work. And that's an amazing thing! It's not only freeing people up to do the creative work, but it's also providing the scaffolding to have a creative process. So, that's really neat.
Galen Low:I like that. And actually, I think even just to Oh, this is a nice round out to this conversation. Something you said earlier was about like the human skills, right? And in some ways, like that creative process, this collaboration that we are managing, yes, it's a project, but fundamentally it's human collaboration. And it's the human aspect actually, that becomes the value, right? The thing that project managers, are being paid to do. The thing that they're qualified to do. And maybe not everyone doing the job today feels like they're, qualified for this sort of human collaboration sort of thing. But in tandem with the tools, how can project managers just learn more and train themselves more around this sort of human collaboration, the human skills, right? The problem solving, the nuance stuff, like rather than go and, learn whatever a new methodology and try and implement it, which is probably fine. But also could that time be spent doing something else that sort of builds up their confidence in human interaction and collaboration management, and if so, what might some of those things be?
Kelli Korducki:Well, I think there's just a number of ways that people can work on building their soft skills. And I feel like I know the Digital Project Manager has written about this stuff before. It's just the soft skills, the core people skills of project management. These have not changed. AI does not change these skills. These have always been the pillars of the job. And the ways to work on building those are the same that they always have been. It's just now I think maybe there's more of an incentive to really double down on focusing on these abilities. Because pretty soon if the technology continues to evolve at the pace that it is currently evolving, there really won't be a lot of wiggle room to not have those just like really strong core skills.
Galen Low:I like that. And actually for me, it's this all ties back into, I don't know, I'm going to say a reframing of value. So, if you're listening to this right now, you're a project manager and the value that you think people see in you and maybe in yourself is around how great of a spreadsheet you have or how fast you are at transcribing meeting notes and creating minutes from it. It may actually be that, there's an exercise to go through to think of yourself as a bit more creative, a bit more strategic, and just picking apart what you do every day, that is the human side, that is the creative side, that is the problem solving side. And maybe just reframing that value for yourself and making that part of your brand, not, hey, I'm a great project manager because my spreadsheet rocks, my Gantt chart rocks. That's all important too, but the thing that generated that, the thing you said earlier, right, the making of a thing, like that process of creating something that has never existed before. You know, arguably, yes, your Gantt might be great, but the thought process that created it is a thing that's really valuable. You probably had to talk to a lot of people to understand how long things take and what depends on what, and what are the key dates. And in so doing, probably had a lot of, navigated a lot of human dynamics of people not wanting to commit to a date or not wanting to share all the information or being too busy for you. And that's the value and that's a skill that, won't to be going away from the profession anytime soon.
Kelli Korducki:Or from any profession, like these are the things that people do. This is why you can't replace people because only people can be people and with people skills. Right? But it's great. Like it, in a way, it makes you realize what human beings bring to the table of the work that we do with each other. It's like maybe simple and corny, but it's nice to be reminded of that.
Galen Low:Absolutely. It's nice and corny.
Kelli Korducki:It's nice and corny.
Galen Low:And you know what? I could probably leave it on a very warm human note, but I'm not gonna because I'm going to pull the rug out under us. We've been talking about large language models. Don't worry, they're taking tasks. You're still human. You're still creative. Your jobs are safe. But actually, looking far down the line, what is the thing in AI in the future of work that might change that? Is there something that beyond large language models, comes up in your conversations of okay, well, when that happens, yeah, then we're all out of a job. Is there that thing looming over us? Are we just keeping ourselves happy for the time being, even though, our professional doom is somewhere, down the garden path?
Kelli Korducki:So people are very, I have asked this question of the best, most qualified people to give a definitive answer. And they don't have one, which is an answer. I mean, is it a few, I think it's like a, you guys should know the answer to this question.
Galen Low:Oh, yeah, okay, no, I didn't think of it that way.
Kelli Korducki:They're like we don't know. I hope it's gonna be alright. So the answer is we don't know. And we're just gonna have to roll with it. But, I guess my optimistic way of looking at things is that by virtue of being human beings who just, we possess just certain intelligence that machines don't have and presumably that machines will either always not have or always be less, have less of, that we'll always be able to have kind of that advantage and, stay ahead of the curve. Who knows what will happen next?
Galen Low:Who knows indeed, but it's better to embrace it than to stick our heads in the sand.
Kelli Korducki:I agree.
Galen Low:Kelli, thanks so much for hanging out with me today. This has been a lot of fun. Great to get your perspective. I know you've been talking to a lot of folks on a lot of different industries about this, and I just appreciate your insights.
Kelli Korducki:Yes, of course. Thank you so much for having me. It's really a lot of fun.
Galen Low:All right folks, there you have it. As always, if you'd like to join the conversation with over a thousand like minded project management champions, come join our collective! Head over to thedigitalprojectmanager.com/membership to learn more. And if you like what you heard today, please subscribe and stay in touch on thedigitalprojectmanager.com. Until next time, thanks for listening.