Just buying the tools and the technology doesn't really work.
SPEAKER_01Are you finding that people are far behind? Are they surprising you with how much they know?
SPEAKER_00There is a wide variation in it. And I think a lot of it comes down to people's individual personal AI behaviors and how that actually filters into their work.
SPEAKER_01How important is the introduction of these like consideration agents, such as the assortment agent and the design perspective agent, in these processes that you're generally working with?
SPEAKER_00So agents typically now are key to figuring out new workflows. And key thing is really how do they collaborate with each other and then actually also with people? The sooner you can get doing something, the better.
SPEAKER_01Jeff Gibbons is a partner at Board of Innovation and a trusted voice in AI transformation. He works hands-on with the world's biggest brands, helping leaders rethink workflows and amplify human skills in the age of AI. Welcome to Using AI at Work. I'm your host, Chris Daigle. Each week we'll be learning how today's business owners, entrepreneurs, and ambitious professionals are getting more done with smart use of tomorrow's tech. Let's get started. Right now, every business leader is asking the same question. What are we going to do about AI? If this is you, ChiefAIOfficer.com has the answer. We give you a simple path forward where we provide executive and team training so your people know exactly how to safely use generative AI in their day-to-day. We also manage the deployment and implementation to make sure tools actually get adopted and deliver results. And we'll also guide company-wide transformation so AI becomes part of your operating system, not just another shiny object. The companies that act now will increase productivity, cut costs, and grow faster than their competitors. Those that wait will get left behind. So if you want to make AI work in your business, visit chiefaiofficer.com and see how we're helping companies of all sizes finally get results from AI. Welcome everybody to another episode of Using AI at Work. My name is Chris Dagle. I'm the host of the podcast, and I'm uh doing this podcast from a Regis cubicle in Charleston, South Carolina, as I uh have been on the road a ton. But I'm very excited about today's topic and conversation with Jeff Gibbons. The reason that I reached out to Jeff was at Chief AI Officer, our parent company, we've been particularly interested in this concept of AI transformation. What is going on with companies who have said, okay, we're ready to answer the question about what are we going to do about AI? Um and with the work that Jeff's been doing with and with the caliber of clients he's been doing it with, uh, I'm particularly excited to extract as many insights as I can about what he's hearing from uh business leaders, um, their level of preparation, their level of understanding when it comes to generative AI and their business. So uh let's dig in. Jeff, before we get started, I think it's important that everybody kind of understands your qualifications and why I sought you out to be on this podcast. So if you don't mind sharing just a few minutes of your background, um and we'll uh we'll start mixing it up.
SPEAKER_00Yeah, awesome. Thanks, Chris, and thanks for having me. So um, yeah, I'm Jeff Gibbons, based out of uh based out of New York today, not Charleston, sadly. And I have been for the last nearly 20 years working with companies on strategy, growth strategy, innovation, and then more recently in the last few years, specifically around AI transformation. So over the last couple of decades, I've worked with companies across all different industries in financial services, in pharmaceuticals, consumer products, uh looking at how do they actually invent new growth effectively and launch things most successfully. And in the last few years, I've been working with companies like Walmart, like the Coca-Cola Company, Nestlé, MasterCard, so some of the largest companies in the world around how do they actually take advantage and develop more competitive edge in the age of AI. So not just about how do we automate or do the same things a little bit cheaper, faster, but how do we actually use AI as part of our competitive edge? And so a lot of work on strategy, on actually building custom solutions, also actually figuring out the human side of transformation. Uh and I think that's actually where I see the biggest opportunity today is that so many companies in the last few years have gotten very excited about AI. They've launched some experiments, they've bought some tools. And I think what a lot of companies are finding is that just buying the tools and the technology doesn't really work. Uh what you need to also think about is how do we actually reinvent the work we do? And maybe there are things we couldn't even do before rather than just automating the old way of working. And then the second piece is actually how do we actually get humans and AI to work together effectively? That's something that a lot of companies will invest in things like prompting and how do you use the tools, but not actually in how do you actually collaborate, how do you evaluate the work of AI, how do you actually understand thinking critically about what the AI should do as well? And that's a lot of the focus of my work more recently. And I actually just published a book recently about that's called Critical Intelligence, all about how to strengthen human thinking in the age of AI. So, how do you actually understand what's the role of critical thinking in a world where you can get AI to answer any question? What are the right questions to ask? How do you actually understand human AI literacy, understand how you should work with AI, and how do you actually evolve your own skills as an individual as well, given that AI is constantly getting better at every single thing, apparently every day.
SPEAKER_01Jeff, I'd I'd love to be able to include a link to the book in the show notes for sure. I'm particularly interested in that. Um just off of that introduction, I've got about four hours worth of conversation that I can have with you because I've got so many things to unpack. Yeah. Um, but you know, why don't we start here? At the time of this recording, there has been uh a lot of chatter around a report that I think was a combination of MIT and Harvard, talking about how 95% of AI pilots, at least at the the enterprise level, weren't weren't seeing ROI. Yeah. Um you are you're obviously familiar with that report. Uh what's your position on that?
SPEAKER_00Yeah.
SPEAKER_01Yeah, I would imagine. What's your position on that? Like what are your thoughts around that? And and is that what you're seeing, a 95% failure rate?
SPEAKER_00Yeah, I mean, I think it's interesting. So I have seen the report, and actually I feel like a lot of people who've talked about the report haven't necessarily read it um and like actually looked at what was in the 95% too. So the 95% was actually that 95% of the pilots that they studied in it were not showing top line revenue impact. So um I think partly that's about the fact that a lot of people are not even trying to achieve top line revenue impact. They're mostly focusing on efficiency and cost savings. I think what I see at companies is that AI is somewhat effective at the automation and efficiencies, but it's not anywhere near as much as people promised a few years ago. Or and so now companies are in a bind where they've basically made a business case around AI is going to lead to automation and efficiency. So they've made promises on that. But what they're also finding is that it actually adds new work to their plate and actually creates more complexities and new opportunities that they also need to manage in parallel. And so, really, the challenge is actually not just how do you implement AI, but how do you implement the technology, get your people to actually use it effectively, and then actually take advantage of all the things that you can do for the first time and that you have to do. So, let me give you a couple of examples there. So if we think about somebody who works in marketing, AI can be great to actually automate some of the parts of how you run a marketing campaign, right? You can get it to generate copy and images and massive time saver. Awesome. That's great. Uh well, I think what marketers are discovering now is that while that's true, you also now have to think about how do I market to agents as well as humans? How do agents actually make decisions? Um it's actually different to humans, and it's not actually entirely rational either. There are actual cognitive biases in studies that are coming out just in the last two months around how do agents make decisions and how are they weird and idiosyncratic in the same way that humans are and in different ways. And so there's so many more things that you have to do as a marketer. Um and also you can't just sound like samey AI content as well. So there's like so many different challenges with it. And so, really, it's it's really like a human and organizational challenge of how do you actually figure out what are the new ways of working, what are the new things you need to do, and then build AI and human systems around that.
SPEAKER_01No, I hadn't even it's obvious when you say it, to me at least, I hadn't even considered now we're not just exclusively marketing the humans. We need there's the landscape is made up of agents who are seeking information, and we need to make sure that our marketing copy and the structure of our content is also not just appealing and authentic to a human, but satisfies what an agent is looking for. That's fascinating. I hadn't even thought about that, but makes perfect sense.
SPEAKER_00Yeah, and so if you think about the MIT study, they found a few patterns in what they were what they were looking at. So um they found, for example, that uh a lot of a lot of the work that companies were doing was just automating existing workflows. So in 79% of instances, they were just automating existing workflows. That's great, but that doesn't actually account for the fact that there are new things you need to do, um, like how do you figure out a new workflow for marketing to agents? And also there are the way that workflows are built today is based around human inadequacies and old systems, and you could actually improve them too. So that process of reinventing your workflows before you install the technology is often a step that companies miss.
SPEAKER_01So how how does a company go about that? That step of, okay, we've been doing it a certain way, let's don't, and it's it might be dysfunctional or inefficient, let's not automate the dysfunction of the inefficiency. Let's re-evaluate the process as a first step and then take that optimized process into a AIFACAM.
SPEAKER_00Yeah, sure. So let me give you give you an example from um a client I've worked with recently. So we're working with a company that's a retailer in the fashion space. And so we were looking at how do you actually integrate AI into the fashion design process. So for this retailer, they were looking at roughly a year-long time frame from I want to come up with an idea for a piece of clothing to it actually being on a on a shelf, uh, on a on a rack. And so we started out by mapping out all the different stages of that, and it's many stages, interlocking functions from merchandising to working with suppliers to design to all sorts of things. And it's a very complex process, and there's a reason it takes a year, right? There's a lot of handoffs, a lot of testing that has to happen. But what we did was rather than just inserting AI into different parts of that process, we figured out what were the fundamental outcomes that they needed to achieve. So to as an example, you need to figure out what is the item or the garment. You need to figure out is that actually something that people would want to buy? Is it stylish? Then you need to figure out does it actually fit into a line? And does it make a cohesive line? Is it right for this brand? Um, and all of those pieces in between. So we identified all of the different outcomes that would need to be achieved at different stages. Could we find a supplier for it? Is it an acceptable price? All that kind of stuff. And then we designed a new workflow that would actually still achieve those outcomes, but do it in a different way and actually in a different order. So one of the things we actually found most useful was to integrate simulation of consumer feedback very early into that workflow, which was not something they did because it would be very time consuming. If you had like a hundred different ideas for address, it would be super time consuming to go and ask people about a hundred different designs for address. But we built a synthetic persona of a consumer, and then you've introduced a new simulation into that process. We then actually built an assortment agent. So the assortment agent would go through the process of saying, okay, we've got 30 designs. How could we achieve the same level of consumer appeal, but in fewer garments? So how could we have 20 garments that would actually satisfy just as many people and not have to make 30 different things? And then we had a creative director agent that would actually look at, well, this design is really good, but is it right for this brand? Maybe it's right for that brand instead. So all of these things were tasks that didn't really exist in the old process. But by implementing and building agents around those, we were able to streamline the process and actually de-risk the design process too.
SPEAKER_01So there's uh a couple of things I want to just make sure everybody that listening that's listening is clear on. They took a process and they broke it into smaller chunks. And at each chunk, they re-evaluated is there ways, A, is the process effective and efficient? Can we re-engineer this process leveraging AI? I think it's brilliant to create those synthetic audiences, the synthetic personas, to introduce something that would have been time and cost prohibitive in the old process, but at least in theory, it sounds like it would result in each step of that process having a much more precise and likely outcome towards the bigger goal. Awesome. Now you mentioned agents a number of times. Uh, how important is the introduction of these like consideration agents, such as the assortment agent and the design perspective agent, in these processes that you're generally working with, not necessarily exclusively with this client. Is that a step that you guys are introducing at each phase regularly?
SPEAKER_00Actually, it's funny actually, when we started working on this um this fashion business, it was almost two years ago. No one no one really talked about agents at the time. At that time. And we so we were, I guess we were building agents, we didn't really call them agents at that time. Uh we were just thinking about what are the things that need to get done and how do we build a specialized part of the process to do that effectively. And I think so. Agents typically now are key to figuring out new workflows. And the key thing is really how do they collaborate with each other and then actually also with people. So there's a lot of conversation you'll see about agent architecture and how do you and there's all sorts of complicated diagrams on LinkedIn that people will post, which some of which I'll be used, I'll be honest, I don't understand. But what you never see on those diagrams is people, right? And so one of the things is actually how do humans interact with those agents? What are the ways that humans fit into those workflows? And so thinking very deliberately about like a human AI workflow rather than an automation workflow. So it's one thing to have a way for you know five different agents to collaborate, but then you also need to know if I'm a human, which of these agents should I trust just to do their thing? Which do I need to double check their work? And what's the role that I should play? Am I slowing them down if I if I get involved? Are there issues that we need to flag that need human review and all those types of questions? So that's really more of the sort of like human AI design that needs to happen within the process of designing new workplace.
SPEAKER_01So it seems like, and we're gonna move on, but I'm particularly I'm an operator. I understand businesses that have clear and clean processes operate more efficiently. So I'm I'm interested in this particularly. In the initial, I guess, uh identification of here's the scope, here's the ideal outcome, oh, let's chunk it down. How involved are the process owners or subject matter experts inside the company participating? Are they collaborating the whole time with you guys? Do you bring them in at certain points?
SPEAKER_00Yeah, great question. So um so yes is the answer. And I think both in terms of the actual initial conceptual design, but then actually in using the system that we're building. So I if I think about this fashion design, we actually worked with the designers and the merchandisers. They were part of the kind of testing group for the initial version of the tool. And it was actually critical to get their input. You might think that some people would think, oh, yeah, you're like automating their job. They were actually massively excited to get involved in this because this was the process was designed to actually streamline the things that were very laborious, quite time consuming, and kind of boring. And so they actually had a tool at their fingertips to run the design process much more quickly and give them inspiration. And they could they could run the whole process like five or ten times in a day if they wanted to. And they were able to amplify the amount of input and creative thinking that went into their job uh through the use of a tool that did all sort of the the kind of laborious analysis process behind it. And so they were critical to actually using it and then deploying it at scale.
SPEAKER_01Any tips for making sure, like obviously you don't want to build something and say, hey, test it, and they go, page one rewrite, this is no good. Any tips on making sure that throughout the process there are those uh inputs. We we haven't built it yet, but we're designing the process. How do we get those inputs from the subject matter expert or the workflow owner so that we're not building something and then having them say that's totally wrong?
SPEAKER_00Yeah, yeah, great question. So I think the first step is actually mapping that existing process and getting them to tell you where the pain points are, and actually getting some of their colleagues too as well to understand pain points uh subjective, right? So pain points in that process might be for the workflow owner or for the people they have to collaborate with. Um so that's that's one part of the process is like give their in get their input on the pain points. And if you're gonna start solving those pain points, they're gonna be interested. Uh the next piece is actually before you start putting code down, you actually map out like here's some illustrative visuals of different stages of the journey, here's a sort of a process map, like giving them something very simple to react to before you actually start building anything. And then as you're actually working, you know, giving them a working prototype of a system, you're getting them to actually try it out on tasks that they're doing in their day jobs. So it's not like an academic exercise. It's okay, so you got that meeting next week that you're prepping for. How is this useful or not in doing that part of the job? And so they're focused on testing whether or not the thing is good for that outcome, not do I like the way that this button is positioned on the page. So they're like, they're looking at the right level of feedback that they should be offering at a very early stage.
SPEAKER_01So, you know, I I one of the first things that I heard when I got involved in generative AI a couple years ago was this media narrative of AI is going to take your job. You and I, I I think Chat GPT is not going to take a job. What you're talking about, though, does transform what the job looks like. How much friction are you getting? Client says, hey, we want to introduce AI, we want a transformation process, we want you to work with our subject matter experts. You go and meet the experts and they're kind of like, I don't, I don't know if I want to help work myself out of a job kind of thing. Are you seeing that type of friction from the the subject matter experts that you're collaborating with?
SPEAKER_00You know, I would say generally no. Uh I think when you're working with when you're working with executives, they are thinking about how can they re-architect the system so that it's more effective. They're thinking about better outcomes. And they're also thinking in their personal work too, like how could this make me more effective as a leader? And they're thinking, and I'd say actually the the bigger concern we hear from people today is I mean, it's driven by a fear of automation, but how do I future proof my career and my skills? So people thinking deliberately around what are the things I need to do and be better at in order to actually be a better leader, to be a better X, Y, or Z. And so it's coming from a place of wanting to keep up, wanting to stay ahead. Uh, but I think it's not about people's jobs. They understand that jobs will exist now, but it's more about how can I be the most marketable, how can I be the most effective, given that I'm gonna have a different job, which will have some new responsibilities that I've never heard of, like how do I manage a team of humans and AI? Like everyone's gonna learn that for the first time. That's a totally new skill. And people are interested in what are the new skills they can take on, uh, not just like how can I protect my existing job.
SPEAKER_01So you're saying generally you're met with maybe even excitement or enthusiasm about participating in this project from the client side, from like the actual doers of the process?
SPEAKER_00Yeah, I mostly mostly excitement, some you know, some shock at some things uh around what's possible. Uh yeah, still, I mean it the capabilities of AR are still evolving and still people get surprised, like, oh, I didn't know it could do that. Like because it keeps changing. And I think there is a sense I see a lot of people being overwhelmed by the amount of change and not being able to keep up with it. I feel that myself. And so I think if you're able to show people within your day to day world of work, here's what's possible and here's why that could help you. Um, they mostly get excited. But I think for me too, there's still a little bit of like terror and exhaustion around all of it, too, just because there is so much to keep up with and so much to think about all at once.
SPEAKER_01You know. A thought that's coming to me is that this is not the sexy part, like the the sitting with the client and saying, Okay, then what do you do? Okay. That's not the sexy part. Could AI do that? Maybe. Would the human be as open to collaboration if it was Jeff's AI agent that was eliciting this information as compared to Jeff sitting across the table or on a Zoom with them? And and then the process part of it, like somebody still has to run the process today as of you know 2025. Yep. So I think for anybody listening that is is uh uh a doomer, you're still gonna need people like Jeff to work with people already in your company to even start the AI conversation about how do we improve our processes. So we're we're not at uh Terminator yet. And I know that you you work with a lot of different industries. Your clients tend to be, I guess, what would be qualified as like upper middle market to enterprise. Is that the size typically? Yeah, yeah, yeah. Okay. They're the the the AI knowledge level of the people that you're dealing with. Maybe the executives say, hey, let's do this thing, here's the strategy. But when you start to work tacally with the departments and the teams, are you finding that AI understanding is all over the board? I mean, I know nobody's a true expert, it just moves fast. Are you finding that people are far behind? They're are they surprising you with how much they know?
SPEAKER_00Yeah, I mean, I think uh there is a there is a wide variation in it. And I think a lot of it comes down to people's individual personal AI behaviors and how that actually filters into their work. So you have some people who are just naturally more experimenters and tinkerers in their life, and outside of you know, outside of work, they're using AI to like generate videos and whatever it might be. And that you have those people, um, and then you have people who are just not that way, and they might have a certain view around things like data privacy, or they just haven't really engaged in the topic. And so there's a wide variety, partly it's attitudinal and partly it's level of experience that you see. Um, so in the same way that there are yeah, sort of segmentations of people who are experimenters versus skeptics, in as consumers, it's the same for for the world of work as well.
SPEAKER_01So uh when I got involved um in like March of 2023, I thought, man, I'm already too late. I saw what was happening and I thought, oh, everybody must be seeing what I'm seeing, right? 2024, I thought the same thing. This is the year. It wasn't now I'm seeing that acceleration, like it went from a trickle to a flood as far as interest from companies who are ready to get off the fence. You've been you've been actively working with clients for the past several years. Have you seen that the the client understanding of it has uh and been and become enhanced over the past few years?
SPEAKER_00Massively, yeah. Yeah, even over the last year, it's it massively changed. I think what's changed is that you know, a year, year and a half ago, typically people were excited and wanted to try. And now they've done some trying and they've kind of figured out some things that worked and some things that didn't. And now they're like, so how do I make it stick? Like that's that's where companies are at right now. They, you know, like the 95% stat in that MRT report. Like a lot of people now have been through one of those 95%, right? Or some of them have been in the 5%, you know, a mix of those things together. And so they've seen some things that have been effective, they've seen some things that have been kind of an abject disaster, and they're thinking, well, like, how do I do all these things differently? Uh I think some of the main things that come through in that are, you know, we've run a lot of experiments, but how do they come together? How do we actually orchestrate and and figure out what's going on and prioritize within that? That's sort of one big problem. Uh the second thing I see is that even if people have kind of gotten on board and they've realized, you know, they've implemented X, Y, and Z tools or they've run new processes, they're seeing that everyone else in the market is doing the exact same thing, right? Like if you work as an executive in a company, you're getting pitched every single day by AI companies and consultants, and so are your competitors. And so they're trying to think about well, how do I differentiate? What are the things that I'm going to do differently within the way I implement it? And that really comes down to how you actually integrate it into your processes. So every company can buy the same tools. It's literally, you know, it's a commodity these days. Uh, data sources, you can, you know, you can use your own custom data. That's obviously an opportunity, especially for large companies, but a lot of the external data is all for sale as well. The the models, everyone can buy the same models. And so, really, sort of how do you how do you actually figure out the integration of humans and AI into new workflows and integrating that effectively, that's really where we see the differentiation opportunity coming today. And I think companies are starting to see that it's it's not just about the technology, you have to think about the humans and how do they work with it. Uh, and that's that's where I see the the mindset starting to shift.
SPEAKER_01So let's let's pivot into that then. If if that's the if anybody listening to this is like, how do we make sure that we we do this right? Yeah, as you mentioned, it's not the tools, it's not that it, but it is your people and how you're integrating the people, not just, hey, let's let's create a different culture, but it's how do I get my people to become those cyborgs, as Ethan Mollett calls it, right? Like they're seamlessly working with these tools. They've changed that reflex of, oh, let me leverage the models for this answer or this I'm stumped. What do I do? Oh, let me go to the models, kind of this new behavior. How do you start that process in a company to to get that cultural uh, I don't know, uh, evolution to occur with their people?
SPEAKER_00Yeah, great question. So I think um the first thing it really starts with is changing the mindset around how you think about AI. So I think a lot of people, their mindset around AI is really driven off the idea that when they started using Chat GPT two years ago, three years ago nearly, some for some people, they thought about it as like it's a chat bot, telling what to do, and I'll I'll see how good it is, right? That's kind of yeah, it's like it's like a chat bot, I'm the user, I'll give it an instruction, and then we'll see if it's any good. And I think what we have to do now is actually start with a mindset shift of actually seeing it as a collaborator. So think about your job as imagine you sat down for work at the start of the day, and there was somebody who was very well qualified and pretty detail-oriented, sat next to you and said, Hey Jeff, I have nothing to do today. I'm right here if you want to talk through anything, if you want to work through anything, if you want to collaborate, if you want advice, if you want me to stress test anything, and they're actually able to offer you feedback on your work too. So you're not just the user, you're actually a collaborator in a team with that person. And if you think about the construct of humans and AI working as a team in a team together, you actually start to see things differently. So once you've started to get people to understand AI as a collaborator, then you start thinking about what do I need to learn to be a collaborator? So in my book, Critical Intelligence, it's really a guide to how do you actually move forward if you're thinking in that mindset. So partly it's about if I can get this collaborator that's working with me all the time to do literally anything I want, I need to think really hard about what are the questions I'm gonna ask it. Right? It's actually that's one that's one of the most important things. Then I need to think about well, how am I gonna evaluate whether or not I'm getting good feedback from this collaborator? And then how can I actually get this collaborator to help me reflect on am I doing the right thing? Am I and so this whole process of reflection and metacognition and thinking about how we think and how we do our work and using AI as a partner in that is really important. And then the last piece is actually around how do you build a sense of how you adapt your skills. So we know that AI is getting better at all these things all the time, right? Whether that's uh reading, writing, it's emotional understanding, it's generating images, it's generating videos, all those things, even using websites and software now. AI can do that for you, right? And so if you think about that, you need to think about in my work, what are the skills that I should really cultivate that AI can't do, right? And how do I actually amplify those? So in the book, we we go through the example of an architect. So an architect, they have certain skills that have what we call like a short half-life, right? So it's how do I use certain CAD software? How do I understand the building codes and regulations, all that stuff? AI is really good at that now. And so you don't really need to focus so much of your effort on that. Um, what you should focus on is things that have a long half-life. So spatial awareness and understanding, understanding client needs, how do you actually build relationships with clients? How do you have a strong feedback process with your clients? And how do you actually be best at those things, given that like the knowledge of individual codes and materials are going to be kind of automated away, you can focus your time and efforts on the things that are not going to be automated so easily, and that's how you future-proof your career. And then getting, you know, once you've gone through that, then you can actually start to build programs where you implement it. So we do programs where we actually get teams to build agents that support them in their work, uh, identify new workflows together, build tools, identify pilots, and then you start to start from the mindset and the understanding and then put it into practice. And that's really where the change actually starts to happen is when people put it into practice rather than just reading about it. And I think the key thing is that it's not about prompt engineering. Like the, and that is a great example of what's going to go away, right? So even the way that you prompt ChatGPT now is different than the way you should prompt it like a year, year and a half ago. So a year and a half ago, everyone did all these prompt engineering courses. That stuff is kind of out of date now, right? It actually kind of figures out the right prompt for you if you work with ChatGPT 5. And so it's thinking a step ahead of what should I be focused on from a learning perspective that I'm that I'm gonna need two, three years from now and setting up for that today.
SPEAKER_01You know, one of the things I I just was with uh a YPO group in Kansas City and yeah, two days ago, and kind of the way that I positioned it was as long as you know what your question is, how you structure the question isn't isn't important. What's more important is you really giving it the context around that question. And kind of what I was instructing these execs to do was just turn on the dictate button and talk to me. Don't like try to talk to AI. We'll let it record on the background, but tell me what what's the problem, what's going on, right? And that seemed to get fantastic results as compared to me teaching them, you know, a persona method or our reverse prompting method of prompting.
SPEAKER_00Totally.
SPEAKER_01So I think that's you know, one of the things that you mentioned Yeah, go ahead.
SPEAKER_00Yeah, no, no, I love that. And I think that's that's a really good example of like thinking of AI as a collaborator. Like talk to it like you were talking to a colleague, right? And and that actually will be most effective rather than thinking of it as like a piece of software that you have to read a manual for.
SPEAKER_01Yeah. So for those of you that want to try that, don't look at the screen if you're new to this, don't look at the screen and think, uh, just kind of turn it on and think out loud. Because it's we want it needs to be that like truly natural context. Don't get intimidated by you know, one of the things that you mentioned about this the architect role, right? It seems like certain um professions have uh characteristics about the personalities that would do those like a like an architect, like a coder, let's say, right? However, if AI is going to be doing the heavy lifting of the coding, that personality that was drawn to perhaps the the the logic and and everything that was required to have clean code, like that personality, they may not be able to exercise that proclivity the way that it was required pre. So uh do you maybe see that some of these professions that that because as you mentioned with the architecture side, it sounds like that would be if if AI is handling the the the technical side of architecture, maybe somebody who's sales inclined could end up moving into architecture because they get to use those people skills way more than the technical skills. Do you see that that might be like we start to have a different avatar of who follows certain professions?
SPEAKER_00Yeah, it's a really interesting question. So maybe to take the the coder first, I would think about it in terms of and it's actually, you know, I was listening to a podcast recently. It's very sad. So many people have done computer science degrees and are finding it hard to get a job now. And I think what's really interesting is understanding the difference between computer science and programming. So computer science is actually about how do computers think, how do they how are they architected, how are they designed, how are systems designed. Coding is like a very executional task within that. And so those are two different things. And actually, what those people are finding, I think, is that the executional task of programming has been automated, but the understanding of like how systems and architectures are built, that's the stuff that vibe coding platforms get wrong, right? So you hear all the time that like you can vibe code a prototype, but then it can never scale, right? Because actually it's not built to scale and it actually is completely weirdly built and architected. And so if you think about that person who loves detail and actually getting in and producing clean code, they can think a level up around clean systems and clean architecture, and how do they really understand what's the right way to architect this system, not the individual lines of code. And so you could have that same kind of uh proclivity could be kind of catered for, but it's sort of working at a different level. And so a lot of those people who you know who were taught to program, I wish that more of their instruction in university was actually around understanding computers and computing and systems rather than the executional task uh of programming. And that maps to that's that's the thing that's gonna be really useful. Like that's actually gonna be like very, very highly paid too.
SPEAKER_01So for anybody listening that is concerned about like how is my career gonna change, going back to the architectural example, it was the spatial awareness, right? Like that's the human sitting in the room and having the emotional experience of when the sun rises going through this particular you know uh structure. Like that's not something that so what I would suggest anybody listening to this who's uh wanting to stay relevant in their career, distinguish the the execution side of things versus that that uh I don't know what you call it, the authentic systems thinking and understanding the sort of fundamental concept behind something, yeah. Yeah. Interesting. Hey, so Jeff, what is uh we're we're kind of getting to the end of this. What would be your advice to a non-technical business professional who's listening to this to kind of get a grasp on things, like maybe maybe even see around corners, like what should I be doing today? What would you what's your advice for somebody like that?
SPEAKER_00Yeah, my advice for somebody like that would be um thinking about not understanding all the new tools and technologies. It's a waste of your time because it's gonna change all the time and you can't keep up with it. I would think about it in terms of do some reading around like how AI systems work, how do they think differently to humans? Um, this is one topic that's covered in the book, like this under fundamental understanding of how AI works on pattern recognition and how that's different to human cognition. Like, understand like what are the key differences. Um, I would say that would be the first thing. I'd say think about what are the skills that you have that you think you're best at, which of those things are prone to automation or not. And then I would think about what are some things in your work and or life that you want to start building tools around? Because it's the it's really the doing and actually putting it into practice, building agents, uh, whatever it might be, that actually helps you to get a real understanding of what works, what doesn't, and how you can fit it into your working life. Uh and I'd say it's the sooner you can get doing something, uh, the better.
SPEAKER_01Well, I'm excited for your book. Your book has been released at this point.
SPEAKER_00Yeah, that's right. Yeah.
SPEAKER_01Awesome. Okay, well, I'm gonna uh again, everybody listening in the show notes for sure. But one more time, Jeff, if you don't mind sharing the title of it.
SPEAKER_00Yes, uh Critical Intelligence Strengthening Human Thinking in the Age of AI.
SPEAKER_01And we can find it on Amazon and bookstores and all that else. Okay, awesome. I'm actually gonna get a copy of it today. Is it on audio yet?
SPEAKER_00Not yet. Options, but yeah, not yet. Coming soon, maybe.
SPEAKER_01Well, congratulations on the release of the book. Uh and if it's diving deeper into as I mentioned when we first got started with in your intro, I was like, man, there's so many things I want to unpack. If I can do that with that book, I'm eager to get a hold of it ASAP. Awesome. Yeah, thank you so much, Jeff. I appreciate it. I know how busy it is in the life of an uh an AI professional in high demand, and for you to take some time to share with me and the audience. And just on a personal level, I'd love to stay in touch because um I've learned a lot from this episode. And I think that there's just not enough, like where can you go to learn this type of thing that you and I are doing? It's not a lot of there's there's no magazine, right? Um, and I think that as I make some discoveries, I'd love to share them with you and and yeah, vice versa. I think that there's there's no competition in this space. Obviously, there's it's so much blue ocean for people like you and I. Indeed.
SPEAKER_00Yeah, no, no, absolutely, and likewise. Look forward to keeping it. Awesome.
SPEAKER_01Well, thank you so much for being on the episode, everybody. Um, I hope you got as much out of this episode as I did, and uh we'll obviously see you on the next episode. And Jeff, thank you one more time. Thanks, Chris. Thanks for watching. Bye. Thanks for tuning in to using AI at work. Don't forget to subscribe for more conversations about how to use AI at work. And a special thank you to our sponsor, Chief AI Officer, for empowering businesses with AI education and training. Visit their website for free AI readiness assessment and AI strategy guide to help you get started using AI at work. It's www.chiefaiofficer.com. Follow us on Twitter at the handle usingAI at work, and visit www.usingai at work.com for free resources to help you harness AI in your role.