How to Stay Competitive in the Upcoming AI-Driven Workforce

Dayle Hall: 

Hi. Welcome to our podcast, Automating the Enterprise. I’m your host, Dayle Hall, the CMO of SnapLogic. This podcast is designed to give organizations the insights and best practices on how to integrate, automate and transform their enterprise.

Our guest for today started as a robotics engineer, the first one of these, we’ve never had a robotics engineer on this podcast today. And then he decided to shift careers and jump into the automation space, which is a massive space. So we’re going to have a good conversation. He’s helped start-ups grow through tech. He’s into AI and analytics and an automation leader for about the last 20 years. He’s got a ton of experience helping companies gain strategic advantage, not just improving their IT infrastructure, but strategic advantage and drive business decisions using data. Please welcome to our podcast, the senior director of AI, analytics and automation at McKesson, JD Plagianis, welcome to the show.

JD Plagianis:

Thank you, Dayle. It’s a pleasure being on your podcast today. I’m looking forward to the questions and what your audience might have for us. Is your audience engaged in any of these?

Dayle Hall:

Only post, afterwards. So engagement comes when we post, then we get other comments. And in some of the platforms, you can make comments. 

JD Plagianis:

Gotcha.

Dayle Hall:

Okay. JD, it’s great to have you on the podcast. Again, you’re our first robotics engineer that we’ve had. We’ve had some different people, different backgrounds. That’s the first one, so I’m looking forward to this. Before we jump into some of the questions that we have around using AI in start-ups and empowering people, because that’s the topic for today, first of all, just give me a quick couple of minutes on your background, how you got into this role at McKesson and how does one move from robotics to automation.

JD Plagianis:

It’s a great question and it’s one of those life moments where the world just hands you a decision that you have to make. In my case, the world took a decision away from me. I was happily doing robotics and moving up the ladder and then I found that I had a noncompete clause and that I couldn’t work on robotics for two years after leaving a company. So I had a pivot moment. And it was good for me that the company that I joined had other things to do. It was an engineering brain trust, but they had a manufacturing space as well. And that allowed me to pivot and get exposure to some Six Sigma. 

That whole lean mindset, I really enjoyed, and I got to use my software background to show up and just automate everything under the sun. It was very empowering. It felt very good. I made a lot of people very happy because I took a lot of menial work away from people. But it also created a dilemma for me because I realized this early in my career, it was so easy to automate roles that people had spent years and years developing and doing and the amount of time that it took them to upskill into a new role was about the same amount of time that it took me to automate that next role. And I started to get some heartburn because I was worried if I could do this, and I’m just a guy, right, there are plenty of people out there smarter, faster, better than I am at this stuff, we could create a permanently unemployable class of people because they could never get up to that faster than somebody could automate it away. 

And so that was the next big moment for me, where I decided to step away from the automation space in general and take my software and my business knowledge and leverage it to start down more of an analytics path. Obviously, automating things generates a lot of data. And I got to see firsthand at large corporations that metrics, whether they’re good or bad, do drive behavior, which would be good or bad in relation to the metric. And understanding how that actually influenced business decision making led me down a path where I could use data to help influence not decisions about bottom line or cost decisions, but that top line number as well, business development strategy kind of things. And it was so easy, there was so much opportunity in the space and I thought, this is perfect. So I had my first major paradigm shift as an individual contributor from solving problems that brought down costs or operational excellence kinds of problems to business development and raising revenue. That really transformed my mindset and got me into the start-up space. I couldn’t resist at that point because there was so much opportunity out there. As a young man, I had time on my hands and a passion inside to get some great things done. 

So I went into the start-up space. And that’s where I had my second major paradigm shift, which moved me from individual contributor to leader of people because there is not enough time in the day when you are in a start-up to get all the things done. Even if you try to automate stuff, there’s just always more to do than you can do and prioritization becomes a big problem, but what really matters is finding and motivating a really amazing team to become high performing. And that became my new passion in life and it’s really served me well for the rest of my career as I’ve moved through leadership companies helping come in, transform teams to become high performing.

Dayle Hall: 

I really liked that as a progression. Obviously, that was, from a career perspective, finding out that you couldn’t do what you’ve been doing, at least for two years, that’s an interesting- and forced pivot, I’m sure you’ve gone through that. But it’s interesting that now, you really look at the empowering of people, and we’re going to get into that a little bit more, because there is- and you mentioned this yourself, there is this sense that sometimes with automation, without a- and some of these things, there has been this fear for a period of time where people are nervous that it will remove jobs, you almost felt that yourself, but now you’ve moved to back into the space, then probably with a very conscious decision but understanding that there is power there, and it doesn’t always have to make people worried that they’re going to lose their jobs.

JD Plagianis:

Yep. The whole time I’ve been out of the automation space, I’ve still been chewing on the problem. I’ve seen what happens in countries where your unemployment rate rises 15%, 20%, 25%, and you get a bunch of folks who want to be productive, who want to be employed, who want to have that moral connection from being busy and contributing to society, but they can’t, and it generates a lot of frustration, and you end up with folks just finding ways to express it that are probably not good for society at large. So knowing that this is a risk out there, I’ve been chewing on the idea of how do you make a transition? Because you cannot stop automation from happening, how do you try to harness it in such a way that it leads to a more productive future where we have a million tiny conveniences for everybody on the planet and drastically reduced cost of living and yet, you don’t unemploy everybody overnight? How do we get there? And I actually have ideas about how to do that, which is one of the reasons that I work at one of the largest companies in the world, is I’m hoping that as we deploy an AI strategy here that makes sense, other companies can come and copy paste that strategy when they realize they need one and they can see firsthand what works, what doesn’t work.

Dayle Hall: 

Yeah. No, I like that. So we have two topics that we’re going to cover today, one is around using AI and these kind of technologies, we talked about integration, automation, but how you use that in a start-up. But let’s keep on this topic for a second, let’s keep on this empowering people, how do they use these technologies successfully. So let’s talk about AI. How do you see AI being used in certain types of roles today? And you just entered it yourself, but do you have AI efforts in place where you are at the augment roles and how are they using it to really help people be more successful, more productive and not as a tool to potentially move people out of the business?

JD Plagianis:

Yup. I think now is the perfect time to talk about this with generative AI being at the very top of the hype cycle these days. So for any of your listeners, if you are not already using AI to scaffold your code, or your PowerPoint deck, or do the basic research before you kick off a project, then you’re in trouble because the next guy over, he is using those things because they’re available and they’re incredibly cheap or free to leverage right now. I remember reading just a couple of weeks ago, there was a study that MIT has published, and I don’t think it’s been peer reviewed yet, but by the time this podcast comes out, maybe it would have, it showed the first significant bump in white-collar productivity in 50 years. We’ve had a pretty steady upward trend over time and never a major jump. 

And the study that they did, it broke tasks down into several stages, and no one used AI to replace their whole job. They used it to augment different stages of their activities where they would end up spending a lot less time doing the brainstorming and the drafting because they were leveraging AI for that. And that saved a lot more time for them in the final editing and say, if you’re giving a presentation, polishing that deck, practicing your presentation, delivering it and tweaking it, that’s huge from a productivity perspective because you could generate several graphs of things ahead of time. If you had writer’s block one evening, you could get generative AI to help you through that. 

And I love actually giving that example because I’ve recently done this as a personal experiment. I’m, in addition to AI, an automation leader. I’m also a failed writer of many, many years and probably have a dozen half finished books on my computer that I can just never get to good enough. But I sat down the other day and said, I’m going to get something to good enough. And I used a certain generative AI that everybody may have heard of to help me write a children’s book about why cats land on their feet. And then I used another generative AI that plenty of people have probably heard of to illustrate that book. I pushed the two together in Word, saved it as a PDF, published to Kindle Direct. The whole process took me less than 45 minutes. And I had a friend joke that I am not really a published author, I’m a published editor at this point.

Dayle Hall:

Semantics, JD.

JD Plagianis:

Exactly. But it lowers that bar for getting to good enough. And when you think about business, it’s really all about get to good enough. It never has to be perfect. It has to be good enough that it makes somebody’s life better, and they’re willing to pay for it. And that can happen in any field. So I think everybody can get to good enough a lot faster, and then polish things up and get 40 hours of work done in just 4 hours or 10 hours. 

Dayle Hall:

Yeah. Do you think that- I’m sure you’re looking at this as your own organization, I’m sure you’re talking to other people around the industry. But obviously, there’s still that sense of will this remove certain types of jobs, which we can talk about, you’ve illustrated that we can use examples of where it augments people’s capabilities, it gets them to be more productive on other things, maybe even cut down some of the 80-hour work weeks to maybe 60 for some people. What do you think the opportunities are for an enterprise in certain roles? Like where do you see the big uplift that AI- whether it’s generative AI, or AI in general, where do you see enterprises being able to really take advantage of these current trends?

JD Plagianis:

That is a really good question and it has maybe two separate answers based on the size of the enterprise. At much smaller enterprises, most of the employees are more directly connected to the business development activities that are happening, they have to react to it immediately, or they’re part of that business development. And so scaling factor there, you can augment your employees and suddenly be able to handle more business without necessarily having to grow linearly. At a much larger organization, you probably have scalable structures setup, they don’t scale exponentially in a lot of cases, but they scale linearly or better than linearly. 

Now the question there is, can you enhance the performance of these departments so that you are no longer bottleneck for your business development in this space? If you imagine a really large company that growth is difficult in this space because there isn’t so much of the pie left to go at with the current business model, if you are able to add revenues without adding costs in any sort of semi-linear fashion, suddenly, you can adjust your pricing models and you can go after other market segments that you didn’t have access to before. And what that allows for is scale enhanced by automation without having to decrease your cost burden along the way, so more people have access to lower-priced things than they ever had before. And I think that’s the big lesson for everybody who’s in a start-up or in a company right now. There are so many underserved needs in the marketplace today that we would all love to have, like even your audience listening right now, probably 10 things that they would love to have that make their lives better, but it’s just out of the price range. 

Dayle Hall:

Yeah. I’m old enough to remember when we didn’t have personal computers, each person on our desktops, I’m that old. But I still think as well as a lot of it is just having the time to improve processes, the time to thinking about how many times have you sat down with your books and go,  I gotta sit down and write that. It doesn’t matter if it’s in the enterprise personally, I just don’t feel like we have the time. So like automation, like RPA, when it came out at the time, if it can take some of the menial stuff, if generative AI can take some of the roadblocks or the writer’s blocks or whatever it is that- and that gives us the opportunity to be more productive, to get more out of our business, and obviously help the enterprise, help ourselves.

JD Plagianis:

Most of these generative AI programs out there, if you look at their training datasets, a lot of it is from business journals that have been published over the last decades. So if there’s a tool that you don’t even know about, just talking to generative AI to brainstorm yourself before you start on a project, you can probably get a lot of direction that can save you weeks of time as you get things kicked off. And that can mean the difference between having a short-term proof of concept or standing up a prototype happening in a few weeks versus happening in a few months versus not getting funded because it’s not fast enough and something else had a higher ROI.

Dayle Hall: 

No, for sure. Do you think there’s cautions, the things that causes individuals and enterprises have to think about? So I’ll give you one example. My daughter who’s 15 is now- she’s a freshman. So she’s asking about, is this the kind of process this generative AI that she could use? So I know a lot of educational institutions are trying to figure this out. And what I said to her is, it’s something to augment who you are and what you know and you’re learning, it’s not a replacement for and please do not put something into this tool and then just slap it on a piece of paper and think you’re done for [God said] don’t do that. So that’s a caution that I have at home. But what about for enterprises as they see this coming in, should they think about cautions or controls like- or is it do we not really know what the potential downsides are yet?

JD Plagianis:

The two spaces that I see with that, one, we’ve already touched on, the just good enough. If you can get to good enough, people are going to adopt it no matter what. You can’t really stop them from using a generative AI program to write their essay. You can encourage them not to, but they might try anyway, right? Because it’s just good enough, just convenient enough. The other side is the trust systems that we’ve built up as a society. We have a lot of licensed things out there. We have a lot of regulations in spaces like my own healthcare, or legal spaces. These are spaces that I don’t see disruption happening too fast in. 

For example, if I had an AI doctor on my phone, that would be fantastic if I were somebody who doesn’t have access to a doctor for one reason or another. Maybe I’m in an unserved or underserved community, or I’m out in the middle of the ocean somewhere, and I just don’t have access, that would be great to have. But those AI doctors, they don’t have the licensing, the oversight, that our medical community goes through, the trust that it’s established. So even if there was another robot that tried to reduce the amount of bad advice that the AI doctor was giving, you still wouldn’t accept it right away. It would take many years, I think, before we move away from our trusted family doctors and the licensing to get there, but it raises a good question because these generative AIs are passing or surpassing the graduate’s scores in a lot of these licensed fields. 

So I really just think it comes down to how fast people can trust things and the fastest way to get there is to prove trust, to build trust. And there are a million different spaces as a start-up or as an enterprise where you can push on this that are immediately bumping up against those trust structures that we have. Try to do it for your favorite Netflix show, just translate that into a different language, and that’s a great way to build trust that it does a good job. Don’t start with trying to translate legal documents from one language to another without oversight of a fluent lawyer, or don’t try to design or architect building foundations in an earthquake zone without a professional engineer looking at it, maybe that professional engineer could use generative AI to speed up their process, but it would not replace them immediately.

Dayle Hall: 

Those two examples I think are great use cases. There are some examples where you could ask generative AI to do something, you could check it. Good enough is your term, which I do like, but then there are some things that you need to still have the human, the expert, the someone that has different experience to go in there. And some of these podcasts that we’ve done, we’ve added a couple of people from human resources, from the people that are on the front end dealing with people. And I think that in those conversations, we’ve talked about responsible AI, we’ve talked about AI ethics, making sure that when you bring this kind of thing into organizations that is, it has some controls, it isn’t something that also goes to continue to promote bias, so bias within AI models, something else we’ve discussed. Are you seeing- would you think the HR teams, the people that are really engaged with the people in an enterprise, is there an opportunity for them, or are there too many risks that this is going to be really hard for them to leverage this?

JD Plagianis:

I think, just like any other industry, HR is going to have to embrace it if they’re going to be able to handle the demands of the future. We are at a moment right now, like fire being harnessed or electricity being harnessed. And you can say it doesn’t work for me yet, it doesn’t work for me yet, but if you keep saying that, you’re going to get passed by completely, and you’re going to become irrelevant. And so embracing it in ways like Excel. Excel was the greatest business tool ever created. We have it on our [Investa]. We use it for everything under the sun. You don’t have to get your IT department’s permission to use Excel for a use case. All you need to do is have a nice license for the enterprise, and everyone has it. And I really see these AI and automation tools needing to become more like that. We had low-code and no-code solutions in the past, which you may be divided on in your audience, where some people think it’s the greatest thing ever, some people think it is actually disempowering. 

But the point now is that we can have these tools and just have a natural language interface conversation to get where we’re trying to go so the whole low-code, no-code problem is almost fully addressed because you’re getting real code written by a computer that can then go back and tweak it or explain it or maintain it in the future to adjust for moving needs. Via reinforcement learning, it can change that code on the fly and keep updating why it’s going to change it or pull previous models, essentially work your whole MLOps pipeline for you. I’m very excited for the capabilities here.

Dayle Hall:  

Yeah. As we’re recording this podcast, it was actually this morning that we launched something that we’re- it’s not fully based just on GPT-4, but it’s called SnapGPT. You’ve had artificial intelligence in the background to help build- to suggest the pipelines to connect your enterprise with applications and data. But until GPT-4 came out, the interface was still a little bit clunky, we’re still going through the tool. But now as we marry that together, our CTO described it as there’s opportunities are for people who are non-developers, don’t have to build the code and he actually used that, our CEO. 

Now our CEO is a smart guy. He’s founder of Informatica, founder of SnapLogic. So he’s clearly smart. But he used the example of now a CEO can say, okay, whatever customers, fan are engaged with us over the last 12 months, you can actually put that request in. And the tool will build the pipelines of data, connect the applications and spit out a result. Where before, he would have to go to sales ops, or customer success ops, or go and find somewhere and then have to build a report and run data and probably use some spreadsheets and pivot tables just so you have a view of it. As CEO, I mean even a simple CMO can probably use this kind of stuff.

JD Plagianis:

This is actually fantastic news. And I think just like Excel is on everybody’s computer, everybody’s going to be able to build these massively more capable pipelines to get stuff done and it’s really going to blur the lines between developer and non-developer. You’re only going to need to call people in when you have something custom. Or in my case, I work in healthcare and healthcare has some of the worst data anywhere. 

I’ve worked in distribution, manufacturing and finance. I’ve never seen anything like healthcare. We have terrible models, which is interesting because it probably should be the most secure and sacred data you would think, that your data is perfect in there, right? Can you even see your data to improve it the same way you can see your credit report, the interoperability that has been around for finance for a very long time? You can go to Alaska, and you can pull out money from your bank, it doesn’t matter which ATM you use. You can’t do that with your medical records right now, even though that’s something we would love to have. You have to teach computers the structures of things right now in this space because it’s not as clearly or cleanly delineated as something that came from finance where the CEO could say, hey, give me this, and it understands what types of fields to go through, how to combine these things because it’s seen it a million times. 

In healthcare, we don’t have good consistent views, but we have models like FHIR out there that describe patient interactions as objects, but not all data cleanly fits into that model the way that it’s stored across things. So there’s still a human element in a lot of this. And I’ve been wondering if generative AI is a good way to address this. The problem is that you can’t put data like this into a generative model to train it. That’s very clear legally that you can’t share patient data with these giant models that could go and spit it out accidentally anywhere.

Dayle Hall:

That’s interesting. Yeah, I didn’t think about our healthcare data as crazy as that. But I can imagine if you’re dealing with it every day, you probably see some of that that I don’t. Moving off talking about how humans and the enterprise can use it, and we’ve had some good examples there from different types of organizations, how they can leverage it. Let’s talk about in general, and we’re going to talk about- because I know you have experienced this with start-ups. 

So thinking about using AI to grow your businesses, one of the things that is very- particularly now with the way the economy is, one of the things that is very top of mind is controlling costs, reducing operational expenses, and that is usually one of the key areas for saying, okay, we need AI. Now what I hear from people like yourself on the other podcast is, you should start with what business problem you’re trying to solve and just saying, I’m trying to cut costs is not necessarily enough. In your experience with people you’ve talked to, what you experienced in your own company, how can someone in an organization, in an enterprise, what are the business models, the business processes they should be looking at and how do they go around identifying those without just saying, hey, we want AI because we’re trying to cut costs? And what are the specific things they should identify?

JD Plagianis:

I want to draw a mental model parallel for your audience in answering this question. Think of AI as a bunch of super intelligent, tireless aliens that landed on the planet and every company out there wants to put a hiring plan together to bring some of these aliens on board. So this works the same way as outsourcing has worked over the last 30 years. There’s been a big push to do it to cut costs or drive scale for your processes, but it has never been perfect on the first try. The pendulum tends to swing and you get some hysteresis around your targeted goal of cost and quality and flexibility because as soon as you’ve outsourced a process, suddenly, you don’t have those subject matter experts in place anymore that can adapt to changes or innovate something in that space. So sometimes you reshore it. And then instead of reshoring it, you find a different way to outsource it. And in the end, the whole pie has grown over time so everybody’s better off and it hasn’t upset society or created some giant economic disaster. I feel that AI, driving through a lot of our markets, if we address it from a use case scenario where we just say, I want to do this one thing, is it better for me to put a human in to do this, is it better for me to put an automation in to do this, or is it better for me to put a human with an automation in to do this, this is where people leadership is going to come into play. 

Historically, you have just thought, I can only get this done with these people, or why are these people doing this at all, we can just automate that and get it off their plates. Those were the two extremes. But now you can ask the people that you have to automate it themselves. Just by chatting with a robot, you can bring in folks that have some of these skills. Obviously, these are going to be pretty hot skills in the next few years. I keep hearing prompt engineer, and I expect shortly, there will be bots that do better prompt engineering than the humans are doing. So if you have a model where you start thinking less about, how do I do the job today, and more about, how do I bring in a process that would scale exponentially with just a couple of folks who are augmented by all of these capabilities, you end up, whether you’re a small business or a large business without that bottleneck, that semi-linear cost bottleneck that we talked about at the beginning of the chat here.

Dayle Hall: 

Yeah. I said a lot of organizations- again, we’re talking about integration automation. AI is that next level, and I know a lot of people look at it as reducing costs. But one of the things that I think is growing in importance is not just cutting the bottom line, but growing the top line. And obviously, there’s a ton of companies, vendors, enterprises that are trying to get you to sign up for their software, use their software because they can help you grow the top line. If you’re in an organization today and you’re trying to look at a certain type of technology or using AI on the bottom line, where do you think the opportunities are to grow the top line? Because what I try and do on these podcasts, JD, is, if someone’s listening to this, I want them to hear something and go, you know what, we should be looking at that. So I think growing revenue is pretty much everyone’s job in the enterprise no matter where you are.

JD Plagianis:

It depends on who you talk to these days. There’s a lot of folks that just say, no, my job is this operational excellence piece. I am going to focus on this thing. That’s somebody else’s job to worry about development. And so I don’t subscribe to that. I’m with you on it’s everybody’s job to grow this pie. And so my advice for your listeners on this would be, if you have an existing process, look at augmenting your existing personnel to make that a scalable process that could accept more business if it came down to it. If you are coming up with a new business development, you should not be thinking of it from the perspective of the traditional, here’s my business, here are my costs with labor and everything and then here’s my bottom line. You shouldn’t be coming up with an AI-first business. That’s what allows us to scale where we are right now because we’re in a society that has more open jobs, at least here in the US, than we have people who are willing to take the jobs. That tells me not that people aren’t willing to work. It tells me that the businesses that we’re trying to grow put people at the center of a process that no one wants to do. 

I talked to a lady recently who was trying to turn around the insurance space. And her idea was, how do you eliminate the bottleneck created by insurance adjusters? Because they introduce bias, they have schedules, they can only be in one place at a time, there’s only so many of them in a geographic region, the answer she had was to do a lot of image processing and take other features that are either freely available or low cost so that you could classify- no generative AI here, just good old-fashioned classification, you could classify an issue into $1,000 problem, a $10,000 problem $100,000 problem, and then send your adjusters where it makes the most sense. 

And you could do all of that simultaneously instead of in a serial fashion. And that would allow people to get paid faster. It would prioritize the work of the adjusters so that you’re only sending them to the places that you really need that expertise. And then it also helps, because you’re getting more data faster, build better models for actuarial tables which help the insurance companies drive down their waste as well. So everybody ends up winning and you end up with a better system than you started because AI-first. It’s not, anytime we have a problem, I need to send a person out to look at it.

Dayle Hall: 

I like that analogy in terms of that kind of opportunity, that kind of role. I don’t think a lot of people would necessarily see that process developing that way. So we talked a little bit about top line, bottom line and AI and how it can help, what to think about processes, business processes. One of the things that I think as a marketer in IT for 20-odd years, there’s so many stats out there around X percent of technology projects fail. There’s always reasons why or so on. And it could be like a new CRM implementation, it could be something like a call center, whatever. There’s so many of these stats saying about how they fail. Now do you think we’re still at the same risk with AI technologies, or because it’s less of a learning curve or it more augments the current people there, is there still a risk of IT projects failing? We asked this question a little bit earlier, which is around how do we have controls, make sure that we’re cautious with AI, but is there still a risk of putting these kinds of technologies and these processes in place in our organization?

JD Plagianis:

I’m very interested in where this goes as well. A lot of vendors have been quick to integrate AI into their product, and I think that can be very beneficial from an exposure perspective. But actually, wielding AI to solve your own problems, not using tools that have AI in them that were designed to solve other problems, that’s where the big trust hurdle is going to be overcome. So depending on your business, if it is not running afoul of the trust systems we discussed earlier, you should somehow get an innovation hub inside your business where the people who are already interested and excited, and there are so many people right now interested and excited about this generative AI stuff, can actually get exposed and play with it in their little spaces. Because the more examples of success that you can find across your business, not just in one department in your business, the more accepting people will be of trying it in other parts and trying it at larger scales. 

I really think a lot of failures come down to leadership, not execution. Oftentimes, either because the sale was bad, you set bad expectations upfront, or you didn’t deliver because you couldn’t meet the expectations, or you did a really bad job of telling people what it was that was going to be solved, like there was a misaligned set of expectations. So the faster you can pivot, the better. And these tools allow you to pivot much faster because you can be more productive. Instead of 40 hours of work and then you get to present and then you find out you need to pivot, you could do 4 hours of work and did 4 hours of work and pivot. And I’m very interested in seeing just how good this generative stuff gets. These large language models, they say, you can solve any problem that can be described in language. 

Well, my problem is that I’m not in two places at once, so my meeting calendar fills up. Can I have some sort of avatar that goes to half my meetings where they just need a decision out of me, and it knows what my preference is, and it knows what kind of arguments would convince me, and it could either make a decision for me or summarize something for me where I could go brainstorm with some of the smartest people in the company without them actually being there and then put together a new idea that I can pitch? These are all great opportunities. If you’re listening right now and you want to generate this thing, call me because I want to buy it.

Dayle Hall: 

I love it. I love it. It’s a good pitch. There’s an interesting thought that now hits me, which is, we mentioned this earlier, which was, my CEO could use the opportunity for this kind of generative AI tool to get the information that he needs. Now, obviously, that’s just essentially like pulling data and looking at it. But what does this mean if you’re in an enterprise start-up or a larger enterprise, and you have technical teams, and the technical teams are pretty much responsible, for the most part, to implement and manage and make sure they’re available, these capabilities, that run the business? What’s the implication for the technical teams? Do you need more people? Does everyone need to understand AI, or do you need more people with more expertise to be successful with this kind of technology?

JD Plagianis:

If you look at most enterprises, the way they’re structured, IT is a cost center, and it’s there to support the business and its execution. And oftentimes, that’s one of the reasons that IT is a target for layoffs. My expectation, because AI is going to enable so many people who don’t necessarily have the skillset that other folks have spent their lives developing, the lines are going to blur and IT is going to become less of a cost center. And I don’t want to say it will decentralize, but the boundaries between the actual business units and the IT structures supporting them will begin to merge where people who were not capable of developing something, like myself, I’m not an artist and yet I was able to illustrate the children’s book that I edited or wrote or whatever you want to call it, but I did, I could never have done that even in my wildest imaginations. I would put pen to paper and it would come out looking terrible and in seconds, I had really good art. 

So that’s why I think we’re going to find some amazing things that come out of the most unexpected corners of our business. And I think it just comes down to people versus the organizational structures. You may be in a database role today, or you may be in a software development role today, or you may be in, say, a governance role today. What are you doing to make yourself more productive and better integrated into solving business problems than you were yesterday? Those are the people that are going to succeed. Everybody else, they’re going to stop getting their promotions, they’re going to stop getting their raises. But there’s going to be so much money being made because all of these businesses are so much more productive, they’re reaching so many more people, the conveniences are going through the roof, they’re affordable for people, the cost of living is coming down. I won’t say- we brought up the moral tie earlier of having a job to the society we live in, feeling good about yourself as a person, I’m wondering if that will start to decouple as well. People seem happy when they have purpose. But can you have purpose and be able to explore new spaces without the fear of not being able to feed your kids?

Dayle Hall:  

There’s definitely a lot of things to think about. In one of our podcasts, there’s a guy called Steve Nouri. He’s in Australia, and he is part of a group called AI4Diversity, I think it’s called. I’m just double-checking that, AI4Diversity. And it’s actually a group that is bringing together people globally to discuss the implications of AI, to have us not just corporate enterprises and people that are creating technology to put the controls in place, but have people who are going to be impacted by it. And I think that actually made- that’s made me feel more confident with what this is going to be for us humans in our daily lives and at work because I liken this to when we launched- when social media came out, we had no idea on some of the negative implications of what was going to happen. And operations are not evil. I just think they didn’t necessarily fully understand where this could go.

JD Plagianis:

They have to move fast enough to keep up with everybody or they die. And that’s the big question with AI. If I replaced my whole labor force tomorrow with AI, or even key pieces of it with AI, what is going to happen? Am I going to be the most successful company in the world, or is it going to be, like the outsourcing example I gave earlier where, oh, you know what, I moved too fast, or I didn’t get the quality or service or cost that I really wanted out of this and now I have to reestablish a team and try again a different way. I really think it’s going to move just slow enough that we will as a society be able to avoid some sort of major social or economic collapse, but it’s going to move us into a whole new world of convenience and joys, I call it the million tiny conveniences. And I really hope for that as the future, I am an optimist. But part of that is having seen the way companies operate. They cannot do something in a single year. It’s always a multiyear learning process, especially with spaces like healthcare and legal.

Dayle Hall: 

Yeah. I’ve been impressed with people like you, people that I’ve talked to around this area, everyone has a level of- they’ve taken on some level of personal responsibility. Not that you can fix or create these guidelines, but everyone knows that, I want to quote Spider-Man, with great power comes great responsibility. And I feel like for the first time, there’s more people thinking about this and really looking at the implication. So that’s, as a father, I think about this for my children, for the future. So that fills me with confidence. 

I know we could talk forever, a bunch of questions we could still go over, but I don’t want to waste everyone’s time just you and I just chit chatting. But I do have a question for you as we come towards the end because some of the things you’ve talked about, the enterprise empowering people I think has been excellent. But you, personally, other than potentially being a published writer of children’s books, publish editor, as you called it, if you look out a couple of years, three years, you pick the timeframe, what’s one of the most exciting things that you think this development around AI, generative AI, or just in AI in general, what are you excited about, JD, when you think of the possibilities?

JD Plagianis:

I think personalization is one of the best things that’s going to happen. Right now, we have standard institutions for everything and they do drive some sort of conformity and normality through our space. But if you want to walk home and you’re feeling a certain way and your house picks up on your body language, then it starts playing a song nobody’s ever heard, but that just perfectly matches your mood, or your kid is just a little too fast in math and a little too slow in their verbal growth, a curriculum can be adjusted in a heartbeat to match and help try different models for learning. All of these little conveniences is the future. And my hope is, we’re going to be able to move fast enough that all of us who had to give something up to prioritize or deprioritize something that was important to us in life, we’re going to get a second chance at taking a shot at solving those problems, or spending that time because we’re going to have so much more free time generated by all of these little conveniences. That’s what gets me so excited.

Dayle Hall: 

Some amazing responses. What did you call it then, a million little conveniences? 

JD Plagianis:

A million tiny conveniences.

Dayle Hall:

A million tiny conveniences. Look, I couldn’t think of a better way to close the podcast at. If I’m out there and I’m thinking about this, a million tiny conveniences is something that I’m going to take with me. And we’ll see where the future holds. But I think it’s an exciting time. It fills me with joy that people are really smart like yourself. Some of the other people I’ve talked to are thinking about doing this in the right way, and not just thinking about how it helps technologies and companies, but the people and helping them in their lives and be more productive. 

So, JD, thank you so much for being part of this podcast. It was a pleasure.

JD Plagianis:

It was an absolute pleasure on this end as well. I appreciate you creating these discussion spaces because, as you’ve said, people are finally starting to talk about things and take responsibility. And I think it’s because they can hear it out there and they’re listening to the voices that are thinking about these things.

Dayle Hall: 

Amazing way to end. Thank you to everyone for listening to this episode of Automating the Enterprises with a million little tiny conveniences. This is Dayle Hall, CMO of SnapLogic signing off. We’ll see you on the next one.