Narrator [00:00:04]:

Welcome to Supply Chain Now. The voice of global supply chain. Supply Chain Now focuses on the best in the business for our worldwide audience, the people, the technologies, the best practices, and today's critical issues, the challenges and opportunities. Stay tuned to hear from those making global business happen right here on Supply Chain Now.

Scott W. Luton [00:00:32]:

Hey, hey. Good morning, good afternoon, good evening, wherever you may be. Scott Luton, and the one and only Kevin L. Jackson with you here on Supply Chain Now. Kevin, how you doing today?

Kevin Jackson [00:00:43]:

You know, this is one of my favorite topics, artificial intelligence. Cause I don't do good with that normal intelligence stuff.

Scott W. Luton [00:00:51]:

Oh, I beg to differ, Kevin. I beg to differ. But hitting aside, this is going to be a home run conversation. I can't wait to hear your insights and expertise along with the panelists we're about to bring in. So, folks, great show. Teed up here today. We're going to be talking about how to unlock the potential of AI and supply chain. Now, of course, we've already seen in so many different ways the immense power and potential of AI in recent years.

Scott W. Luton [00:01:16]:

Today we're going to be talking about some of the key pain points that supply chain leaders in particular have when, as they're looking to address challenges with AI. We're going to be talking about some of the more common challenges when it comes to that implementation and all the solutions that abound. We're going to be offering up some key considerations that teams must account for when it comes to an AI powered supply chain solution. And we got a bright future ahead. We're going to get some predictive looks for what's to come in 2025 and beyond. Kevin, all of that and a whole bunch more should be a great show, huh?

Kevin Jackson [00:01:49]:

Oh, yeah, absolutely. But before we start, I just want to say, don't be scared, okay? Because AI is good. It really is. Just like, it's just a tool that can be leveraged to improve your life and your business. So you better stay tuned. You know, listen to what our guests say. This is good stuff.

Scott W. Luton [00:02:11]:

Well said, Kevin. I would just piggyback on that comment. Don't worry about the fomu. The fear of messing up, right. We learn this is a, you know, global supply chain these days is like an Olympic obstacle course. It's challenging, but we're going to provide some been there, done that guidance and expertise here today to help you get through. So with all of that said, I want to welcome in our outstanding one two punch of panelists, starting with David Vallejo, vice president and global head of digital supply chain product marketing at SAP and Ronny Horvath, partner digital supply chain with IBM Consulting. All right, David, how are you doing today? Great to have you.

David Vallejo [00:02:52]:

Hello, Scott, Kevin and Ronny could be happier to be here.

Scott W. Luton [00:02:56]:

Outstanding, outstanding. Well, great to have you back. And Ronny, welcome to you. Great to have you to the show.

Ronny Horvath [00:03:02]:

Thank you for having me, guys. Hi, Scott, Kevin, David, nice to see you again. We've seen each other in person a few weeks ago, right, so, well, this.

Scott W. Luton [00:03:11]:

Is going to be an incredible digital twin of that conversation y'all had not too long ago. We got a lot of good stuff to get to, but you know what?

Ronny Horvath [00:03:18]:

I don't know. You want to have that conversation, Scott, let's leave that out. It was more private. It was on a party at Sapphire, so we don't want to shed light on what we discussed.

Scott W. Luton [00:03:27]:

All right.

Kevin Jackson [00:03:27]:

I love sapphire secrets. Come on. Sounds like a great television show.

Ronny Horvath [00:03:32]:

Not going to happen. I'm losing my job here. Over there.

Kevin Jackson [00:03:35]:

That's right.

Scott W. Luton [00:03:35]:

What goes on? SAP Sapphire has got to stay largely at SAP Sapphire. But, hey, I digress. So this is where I want to start. David and Ronny and Kevin, early fall, I should say, is right around the corner. It's already been a long, hot summer and so many places. Right. So this is what I want to ask. We've had a lot of fun asking a variety of our guests this here.

Scott W. Luton [00:03:52]:

A lot of our guests have been up to some really cool things over the summer months. I want to start with you, David, and I want to find out one really cool thing that you did this summer.

David Vallejo [00:04:04]:

One of the really cool things we did this year was spending our vacation in a completely different way. Normally we spend planning the vacation and planning out where we stay. This time, we basically just booked in and out flights to Europe and said we just go with the flow. And I like what Kevin said. AI is not scary. We actually leveraged jet GPT to say, where should we go next from here? And it brought us into places. We saw Salzburg, we went into Italy, boats in Mehran, we went into beautiful places in Austria, and it was phenomenal. AI helped us, actually.

David Vallejo [00:04:46]:

Like, how do we spend the day? What should we see? What are the priorities? We're hiking in the Danube valley and things that we have never had really planned, and it was one of the most fantastic vacations. And we had our travel coach and assistant by us every day. So I cannot agree more, Kevin, when you said it's not scary, it can help us, it can help us during vacation as well.

Kevin Jackson [00:05:14]:

Yes. By opening your mind. Right. AI allows you to open your mind.

David Vallejo [00:05:20]:

Exactly.

Scott W. Luton [00:05:20]:

And based on what I heard there, also open your. Your pocketbook or your wallet experience, David. But, you know, kidding aside, I love when we can mix our personal journey with some of the latest and greatest, you know, industry innovations, including technology. And I'm gonna have to try out this approach of traveling by AI. So appreciate you sharing that, David. Ronny, that's gonna be tough to top. As David shared, I pictured all those sceneries and the rivers and the landscape. You name it.

Scott W. Luton [00:05:51]:

Ronny, what was one of the coolest things you did this summer?

Ronny Horvath [00:05:54]:

Yeah. So, first of all, you mentioned summer's around the corner. Please don't, Scott. I live in. I live in Phoenix, so I'm actually waiting for the temperatures to cool down a little bit. Right. But I love it here. Three months, it's very unbearable.

Ronny Horvath [00:06:08]:

But the rest of the nine months are really nice. So I'm looking forward to fall and winter and spring. But what we did is. So, first of all, let me start with the confession. I'm german. That's a pro or a con, depending on how you look at it. I moved to the US in 2017, and ever since I've been here in the US, I haven't really seen a lot from the country. I've been to almost over 30 states or so for business.

Ronny Horvath [00:06:32]:

I've seen many, many cities. However, I've only seen the airport, the hotel, and the client side.

David Vallejo [00:06:39]:

Right.

Ronny Horvath [00:06:40]:

And whenever we were vacationing, guess what we did? We went back to Europe, which is. I mean, it's fun, but I know Europe in and out, right? I've been there, lived there. I've been raised there. And so this time we said, like, no, we want to know more about the US. So he said, look, what options do we have? And the best option we came up with, let's do a road trip. Let's drive from Phoenix.

David Vallejo [00:07:02]:

Love it.

Ronny Horvath [00:07:02]:

To Miami. If I think about that now, I probably won't do it again. But we took two weeks to drive the whole tour. Like Texas, Louisiana, and then Tennessee, North Carolina. We went then to Florida. So it was. I've seen so much. I haven't seen that much in a vacation for a long, long time.

Ronny Horvath [00:07:25]:

Similar than David, however, I did not use AI, and I don't know why. I actually use old school Google and stuff like that. And it took so much, so long time. I probably should have leveraged AI. I just didn't think about it.

Scott W. Luton [00:07:37]:

Well, you know, that sounds like a wonderful, certainly a big highlight of the summer. But two quick observations, and, Kevin, I'm coming to you next. First off, I'm ashamed to say I think Ronny's been to more states than I have been to. And secondly, I didn't even get as much of a honk of the horn, Ronny, as you came through Georgia, so I'm really disappointed. Next time, we'll have to fix that. All right, so, Kevin, David, and Ronny, they had some really cool summers. How about you? What? What's one of your highlights?

Kevin Jackson [00:08:03]:

You know what? Well, actually, I just came back from Grand Cayman island and talk about laying on a beach and enjoying the sign and great caribbean food. Had some jerk chicken. It was just awesome. And the cruise ships weren't in, so there was no crowds. You know, I tell you, that's always one of the best vacations I've had in a long, long time.

Scott W. Luton [00:08:30]:

I can tell.

Kevin Jackson [00:08:31]:

But, you know, I love those road trips, though.

Scott W. Luton [00:08:33]:

That.

Kevin Jackson [00:08:34]:

That was a great option. And in and out to Europe. Wow. You know that? Going with the flow, doing what AI tells you. Hey, I'm gonna have to try that out.

Scott W. Luton [00:08:45]:

Yeah, that sounds like a reality show, by the way, David, we should have sent our film crew and tracked you down and published it to VH one or something. But both of those sounds like great trips. And, Kevin, I could tell that was a great vacation because you told that story, like, with your whole body. I could just see how relaxing it was. Okay, folks, we got a lot to get into here, but I appreciate you sharing all a lot of your personal highlights from the summer, but we got a lot of information and expertise to share with our audience here today. I want to start with this. You know, this ever fast moving world, velocity picks up, it seems like, by the hour. I want to start.

Scott W. Luton [00:09:19]:

Context is really important. That's where I want to start. So, Ronny, I want to start with you, if you would. Let's level set a bit. Tell us a little about yourself and what you do at IBM.

Ronny Horvath [00:09:28]:

Yeah, I've been in supply chain for 17 years or so, and I just love supply chain. It was all with SAP, because I started my career in SAP consulting in various firms, from small firms to now, the large scale firms like IBM. I have been just focusing on the stuff that I really like in supply chain. Everything that moves, everything where I can make a difference in supply chain. We're still replacing paper, and that is something we can debate, because I hear a lot of people saying, no. Supply chain is very innovative. Of course it is. But I still see customers even in the multi billion dollar range.

Ronny Horvath [00:10:04]:

That still have manual processes. And this is what really motivates me to work in supply chain because I can see that difference now at IBM, I'm in the SAP practice and I'm responsible for the digital supply chain practice with SAP, which is everything from design to operate. Right. And I'm pretty sure, David, if you don't talk to it, I can obviously make some marketing here for SAP. So I've been bleeding blue for all of my career, starting and working with different companies, but always with SAP.

Scott W. Luton [00:10:35]:

I love that. Sounds like a lifelong mission to tackle manual processes and make the human factor across global supply chain, give them easier, more successful days. One of my favorite stories in this global technology age we live in. Kevin, I'm gonna get you to comment right after we figure out what David does at SAP. And Dave, tell us about yourself as well.

David Vallejo [00:10:57]:

Yeah, so I'm very passionate about supply chain, just like Ronny has been pretty much most of my career. I started out as a software developer and supporting supply chain software. I kind of stumbled into that, but I got fascinating by the diversity of that domain. And now I run product marketing, which is really thinking about the latest trends, thinking ahead, not just what customers are needing today, but what may they need in three, five years from now. And sort of having that telescope and informing our innovation team of here's what we should think about building across the entire supply chain portfolio that we have. And then, of course, also responsible for making sure that we explain and message what we have, what our differentiators are and help customers making the right decision when making technology choices, but also making choices in their journey and informing them the things that we'll be covering today about being ready for technology, what that means from an organizational standpoint. So product marketing is actually a very diverse discipline that we have at SAP. Working with our customer teams, working with our innovation team, working with the technology platform team.

David Vallejo [00:12:18]:

So super, super, super exciting. And I'm very fortunate that as passionate I am with the subject that I can basically live that out at SAP. So very wonderful.

Scott W. Luton [00:12:30]:

That does sound wonderful. And Kevin, I tell you what I heard there with David. A couple elements that I love. The diverse ecosystem we all know about how that powers the industry forward, whether it's from a bottom line standpoint, innovation standpoint, some different ways we all gain there. And then secondly, that forward looking visionary aspect in that part of the world, man, I'm jealous. I got to travel more and I got to innovate more. David and Ron, you're setting up a high bar. But Kevin, what'd you hear there and what they do?

Kevin Jackson [00:12:58]:

Well, first I want to say, if I prick my finger here, you see it blue, right? Because I was at IBM, I was a worldwide sales executive for mobile wallets and voice at IBM for about six years. But even way back then, we were focused on innovation in business and how to apply these technologies to both the top line and the bottom line. So it's great to see that legacy continuing. And you're talking about paper. We're still using paper, but it's about having the right tool for the right job. Right? So paper is the right tool for some jobs, just like AI is a right tool for some, but it's not right for others. And I think that's one of the things that is going to be important about this conversation, understanding where to apply AI and how to apply AI.

Scott W. Luton [00:13:53]:

That's right. Well said, Kevin. And hey, if you prick yourself, I see blue blood coming. I'm taking you to the emergency room. Just so you know, I'm level setting here. Got to take care of my, my dear friend Kevin L. Jackson. Okay, so, Ronny, some folks may not know out there, we got one of the sharpest audiences in all of global business here at Supply Chain Now, but it may be new to some folks.

Scott W. Luton [00:14:12]:

IBM, SAP have been partners for some 50 years. Tell us a little about that partnership, Ronny.

Ronny Horvath [00:14:17]:

Yeah. One of the reasons why I switched to IBM last year, 2023, I was before in a company that was focused heavily on supply chain in that niche skill. And I built the whole supply chain practice. And for me it was important. I wanted to see something different. But three things were important. Number one, longstanding relationship with SAP, and I'm going to tell that in a few seconds. Number two, it has to be innovative.

Ronny Horvath [00:14:43]:

And number three, the logo has to be blue, like I just mentioned. Right? I can't change to any company where the logo is not blue. So let's start with number two first, innovative and innovation. So IBM has not only consulting practice, but IBM is a technology firm with the consulting arm. And that's one of the biggest reasons why I changed to IBM, because I wanted to be that kid in the toy store that comes in and sees all these wonderful toys that I can play with. Think about it. IBM research with quantum computing and AI. IBM has been in the AI space since 2011.

Ronny Horvath [00:15:17]:

I don't know if you know that or if you remember IBM Watson back then, deep QA won jeopardy against the reigning champion. World champion. Nobody understood what he was. I didn't understood what it was. OpenAI started in November, I think 2022, right? And everybody talks about AI, even my mother who's like 82, which scares me, really. I saw like I need to switch to a company, right? That has like this ground principle and this understanding for long, long term. And IBM is that company because they've been there for a long time, over a decade. And the last but not least most important topic, a good partnership with SAP because my career, as David mentioned, the whole ecosystem around SAP gave me the possibility to be who I am today.

Ronny Horvath [00:16:00]:

If there was no SAP, I wouldn't have a career, even though I never worked for SAP directly. But I worked with SAP all of my careers in 17 years. So for me that was very important. And I looked into that obviously. And IBM being there since, you know, over 100 years or so, they have been partnering with SAP from the get go. 1974 is when they started. I think IBM has like 46 pinnacle, SAP Pinnacle Awards. But David, you correct me here or our marketing team has to.

Ronny Horvath [00:16:28]:

We have four hundred twenty five s four HANA projects signed or so right in the pipeline. We have about 6500 clients that run on SAP worldwide. So SAP is a big, big part of IBM consulting and IBM technology, obviously. And we have about 40,000 SAP technologists around the world, right? And 20,000 of those are focused on s four and transforming businesses because this is the way forward for all of us. So these were compelling arguments. Again, the logo had to be blue, so.

Scott W. Luton [00:17:04]:

All right, Dave, anything you want to add to that longstanding partnership before I get Kevin's comments?

David Vallejo [00:17:10]:

No, I think that Ronny described it perfectly. Right. It's the combination of the consulting expertise of people on the ground that understand the business, but the technology side as well, partnering on the AI front, for example. So a very fruitful partnership. Of course.

Ronny Horvath [00:17:27]:

One thing I'd like to throw in there, David, I think we talked about it in Safire. Christian Klein, we know him, SAP CEO, he said something very nice about IBM one day. It was not that long ago, it was around Safari. He said, IBM finishes more projects than they start. And this was something that really felt good because it means that a lot of projects are in jeopardy out there, right? And most of the time people and customers, NSAP come to IBM, said, please fix it. And the reason is about all these credentials that I told you earlier, we have so many people, we have so much experience in what we're doing. So I think that really went down. Just like.

Ronny Horvath [00:18:05]:

How do you say that in English? Warm butter? I don't know. They're saying, no idea.

Scott W. Luton [00:18:10]:

Hot knife through butter, I think. Hot knife butter.

Ronny Horvath [00:18:13]:

Got it.

Scott W. Luton [00:18:14]:

Butter is never bad, though. Let's just be clear about that. Butter is good on everything. All right, so, Kevin, we heard a lot there about the partnership. I want to reiterate, relationships have always mattered, and we've learned further in the last couple of years just how really critically important that is. And secondly, 1974, who'd have thunk that was the year that gave us Kevin L. Jackson, and it gave us this partnership between IBM and SAP. That is your birthday year, right? Kevin?

Kevin Jackson [00:18:37]:

I believe.

Scott W. Luton [00:18:38]:

Ageless wonder.

Kevin Jackson [00:18:40]:

I'll send you my tithings later today.

Scott W. Luton [00:18:43]:

But what you heard in that partnership there, Kevin.

Kevin Jackson [00:18:47]:

Well, it's like, what did I remember? I think about twelve years ago I went to Waldorf, Germany, and I have a big background in cloud computing. And why was I in Waldorf, Germany? That's the headquarters of SAP. And I was teaching cloud computing to the s four Hana team at the very beginning of this whole transition to cloud and s four honor and to see how things have changed and advanced and how it sort of revolutionized, you know, erp s four Hana working with IBM and infrastructure and professional services team. I feel like I was there at the beginning sort of birthing this whole, his whole relationship. So I'm a proud papa, that's all I gotta say.

Scott W. Luton [00:19:41]:

Oh, I love it, Kevin. That's gonna be a whole new series there. Kevin. All right, let's get into more. So we've really set the table well, especially amongst the relationships and what they're, what the relationships here are driving out across industry. All right, so I want to dive into our key theme here today, unlocking the potential of AI in supply chain. Now look, there's a ton more potential, and let's not take anything away from the real gains we've seen, the exciting gains we've seen in recent years. Now, later in this conversation, we're going to talk more about the ever present, really intriguing topic of generative AI.

Scott W. Luton [00:20:14]:

But I want to start by focusing on AI more generally, especially when it comes to enterprise and supply chain industry. Now, David, Ronny, and Kevin, I want to start with maybe a little tip of the hat to our dear, long departed friend Maslow. Right. Member the hierarchy of needs. So let's talk about what are the key motivators for companies that are exploring the use of artificial intelligence in global supply chain.

David Vallejo [00:20:37]:

Yeah, I think some of the motivators are just, like I explained with a vacation I did, is just curiosity and interest. So oftentimes we hear customers, there has to be something that where this is useful, right? So they don't necessarily set out in a very fully baked use case. But just like, let's try this out. Let's have a pilot, right? One customer said, we have more pilots than United Airlines. Let's just experiment and see what comes out of it. And a lot of that spirit is what we lived at SAP as well when we said, let's have hackathons. Let's just do wild and crazy things without really prescriptive. Here's exactly what we want to do.

David Vallejo [00:21:22]:

That is one key motivator. I think the second motivator is around productivity. You think of all the manual efforts that are going on in an organization and breaking through that last frontier of there's still paper and email and people have to spend time to write summaries and sort of that intuition. Well, this is something I believe that generative AI could probably help with, right? And rightfully so. These are some areas in terms of automation of, you know, we're still a human being has to do a lot of writing and a lot of keying in manual transitioning between physical and the digital world. This is perhaps where, you know, generative AI could, could help. So around productivity, the other is around quality, right? And making sure that, you know, with the huge amount of data that is out there, that we're not missing something. That's why, you know, generative AI can help in some processes to summarizing something.

David Vallejo [00:22:24]:

We're not missing the essence of something. And then business growth. If I'm saving time and effort on the productivity front and getting people out of the mundane manual grinding work, well, then I can deploy the exact same human beings on a lot more meaningful business decisions, right? So helping the business to effectively grow is probably another motivator ultimately. I mean, what we're suggesting companies is do the wild and crazy stuff, right? But also think about how that ultimately provides value, number one, and how that ultimately also scales. Is that something that you're building, something that you can, you know, get out of that little pocket and something that you can deploy across the business in a safe and scalable manner? Ultimately, getting out of that sort of pilot stage is what we are very, very serious about. What we bring to the market we know goes into the heart of enterprises is in very mission critical environments. So that scalability part is for us ultimately a key motivator to develop our software.

Scott W. Luton [00:23:38]:

David, really appreciate your response there. Curiosity, productivity, quality, all big key motivators. And I love your emphasis on wild and crazy experimenting in pilots. That reminds me of Saturday Night Live, that famous skit back in the day. Ronny, would you add anything to those motivating factors for folks exploring AI in global supply chain?

Ronny Horvath [00:23:59]:

Everything was spot on. I can just emphasize that curiosity is what it starts. Or we could also sometimes call it pressure, peer pressure, because everybody does it right now, right? So everybody has to have the feeling I have to be in there, too. I told you, my mom, she's 82, she's using AI, right? And that scares me a bit.

Scott W. Luton [00:24:19]:

It's a little bit scary, but it's exciting. It sounds like. I bet your mom is experimenting with AI and finding out where it fits in most with her journey, right?

Ronny Horvath [00:24:28]:

Yeah. I don't want to talk about my mom here. Mom, I love you, but this is not about you. But I think he's right. The most important thing for our clients that I see out there is productivity. Or let's call it efficiency, right? Because specifically in supply chain, with all the planning that goes around, like demand forecasting, inventory management and planning logistics optimization, there's so much data to consider. And we've helped clients the past decade to automate things, right. With intelligent workflows.

Ronny Horvath [00:24:55]:

SAP came out with something that I still love, and I will ever love it for the rest of my life, called BRF business rules framework plus, where you can model the deepest algorithms as well as rule sets for business rules. And I think those things help clients to automate certain things and be able to handle all this vast majority of data. Right. Because it's just so much. But I think this is where AI is. This is for us, top priority point number one. It's efficiency, productivity. Every client talks about it.

Ronny Horvath [00:25:25]:

Whatever David said, spot on.

Scott W. Luton [00:25:27]:

Outstanding. All right, Kevin, what else would you add there?

Kevin Jackson [00:25:32]:

Well, I really like the emphasis on exploration. Try out these new things. But there's two areas that I like to say. I think AI really comes to the forefront, and the first is data. Ronny was talking about all this data that you get where AI can be used to assess and mitigate risk. Basically, by analyzing data from all these various sources that are collected, it can predict these disruptions, like natural disasters or geopolitical events, or even your supplier issues. This allows companies to develop contingency plans, all of the different options, and to maintain resilience in their supply chain. And second, one of the most important aspects of any supply chain is visibility.

Kevin Jackson [00:26:22]:

And AI enhances that visibility by tracking goods from production to delivery. And this transparency really helps in identifying the bottlenecks and ensuring timely deliveries. It also improves collaboration between all of the stakeholders in this supply mesh.

Scott W. Luton [00:26:46]:

I like that mesh. And, you know, it goes without saying, all of that you described, empower, decisions be made, outcomes to be gained, which visibility is quickly becoming table stakes. I believe we've talked about this before, Kevin, and it's about the outcomes, about what we do with it. We got so much more to get into with David and Ronny and Kevin here. I want to shift gears over to pain. So what specific pain points are those that you're finding in supply chain management and operations? What pain points are folks trying to address with AI? And then secondly, two part question. What are some of the business benefits of an AI driven supply chain? Ronny, pain and benefits goes together. Who would have thought?

Ronny Horvath [00:27:26]:

Totally. And I think I come back to the topic that I said earlier, lise touched upon. Manual and repetitive tasks, right, in supply chain. Again, sometimes you still replace paper. And yes, you're right, paper might be the right approach for certain processes. I agree. However, by replacing paper with technology, I think this is where we can really gain business benefits and value. Because make no mistake, I mean, every change costs money, right? You want to understand what's your return on invest, right? And if you're replacing paper, I'm telling you, this is where we can get the most value out of it, right? But if you think about typical supply chain processes, let's say from the very beginning, data entry, the whole order processing, generating reports, because you need information to understand where you are, where you will be next week, and what do you have to do to avoid certain things.

Ronny Horvath [00:28:19]:

All of this is so highly manual. And this is the biggest pain point for most of our clients, even though sometimes they don't realize it, because I think that there's a misleading fact that bringing in AI can solve everything. So they're looking for the biggest motivator, the biggest change agent for them, how they can bring back most value. But they mostly overlook these little things, these manual processes that take so much time. And by that, AI can definitely help. And the benefits are basically open. They're lying on the table there. Because if you think about it, if I can quote IBM's own generative AI story, IBM has a hardware business, right? They deliver servers, power service mainframes across the world.

Ronny Horvath [00:29:02]:

They deliver the service parts for it, make sure that they operate right. And they had a service level agreement that was benchmarked at like 95, 98%. And they had struggles to meet those service level agreements, right? But years ago, four years ago, they said like, hey, why don't we use generative AI for that because we see a lot is buried in manual processes. We don't even know where we are at times. Let's be proactive about this. And what they did is they took Watson back then, not Watson X. It was all Watson platform. Put that on top of all the solutions.

Ronny Horvath [00:29:33]:

We use an SAP ECC back then. We're now switched to SAP s four. There was an SAP Ariba. We had data warehouses like data sphere from SAP and so on, certain things, but also third party products because IBM was not fully, completely on SAP yet. And what they did with Watson, they put it on top on this data layer and all of a sudden they're just handling Watson to do most of the tasks that were done manually before and they gained huge efficiencies. We're talking 95% of faster response time to supply chain inefficiencies. The one example I always come back to is Europe. Volcano eruption in Iceland.

Ronny Horvath [00:30:12]:

I think it was 2010, I don't remember 2009. It happened, it was in the news, everybody.

David Vallejo [00:30:18]:

Ooh.

Ronny Horvath [00:30:18]:

Luckily nobody, not too many people died. Everything was kind of safe. But then it slowly degraded into a topic where they said like, well, this impacts air traffic, right? And for the human being to understand that and then to translate it to the supply chain and that it will cause problems, it was just way too complicated. And an AI driven solution could have read all these weather data that you had out there. And Kevin, you made that example earlier, right? It could have read that and at least proactively alarm your supply chain folks, that, hey, there's something coming up. 40% of our shipments are air freight. Air traffic is completely down in Europe. We should do something about it, right? And this reaction time, instead of doing it manually, is, I think, what is the biggest pain point and also the biggest benefit that you can gain out of AI?

Scott W. Luton [00:31:05]:

Ronny, I feel like we just got a certification in that last response. For sure. A lot of good stuff that we could have a whole show dedicated to. I would just say, David, anything to add to these pain points and the benefits that Ronny was sharing?

David Vallejo [00:31:18]:

Yeah, I think, I mean, again, the elephant in the room oftentimes is will AI replace jobs, right? And I think the way to look at this is actually the pain point is it's hard to get people to do boring jobs in this current age. And I give you an example, like when we were hiking through the Danube and there was like a beautiful restaurant where we came to and it was like, all right, you have all this outside seating, but in the restaurant there was no waiter, there was nobody. It was a wall of vending machines, believe it or not, that had fresh cake, that had coffee, that had beer, that had fresh pastries and everything. And I talked to the owner who came out and she said, well, we had a really difficulty finding workers and it started during the COVID times and we have started to automate things. And it's the same thing in enterprises. It's tough to get people from the next generation to say, you know what? You're going to stand there for 8 hours and you're going to move the data from left to right and you're going to do this very repetitive task. They don't want that. They want a purpose.

David Vallejo [00:32:24]:

They want to have essentially respect to their human intelligence, to imply their human ingenuity. And so I think AI is all about not replacing the human with a robot, but taking the robot out of the human.

Kevin Jackson [00:32:40]:

I like that.

David Vallejo [00:32:41]:

And so that is actually a pain point for organizations. They still have all these jobs where manual, repetitive things have to be done. So I think that's one of the promises that AI, and generative AI in particular, to get to that last frontier of the knowledge worker, to be respectful to their brain and say AI helps you to make decision instead of you grind along your daily work.

Scott W. Luton [00:33:07]:

Human history shows this, that AI will create far more jobs than it will eliminate. And David, I love a lot of fulfilling, purposeful work. That's what folks are after. Kevin, your response to those pain points?

Kevin Jackson [00:33:20]:

Oh, yeah, absolutely. You know, I'm going to pull on that human aspect, right? And this is all about digital transformation. To be honest, as many industries, companies, the world adopts new technologies, new ways of doing things, they have to drop the legacy infrastructures. So how do you bring the humans along in this constant change environment? Well, it's about training, it's about education, it's about meeting the employees where they are, to take them where they need to be. And so AI can actually address the pain point of improving and making it better for humans as they transition from the old way of doing things to the new way of doing things. And that's the second point I want to highlight. There's going to be organizational resistance in any change. Human don't like change, period.

Kevin Jackson [00:34:24]:

So employees are going to resist AI adoption because they are afraid that they're going to lose that boring job. It may be boring, but it keeps food on the table. It's really important to have effective change management to ensure that your staff are trained onboarded properly and that they see AI as a tool, not a replacement. To enhance and not replace their work. So artificial intelligence can actually provide that education, that insight to the employees. They are all asking, what's in it for me with them?

Scott W. Luton [00:35:09]:

That's right.

Kevin Jackson [00:35:10]:

Use AI as a tool to address the question with them.

Scott W. Luton [00:35:17]:

Love it. Kevin okay, moving right along. David, on this is a related note. When you think about common challenges, obstacles, how organizations are, or how they should overcome some of these challenges we're talking about, and any examples that come to mind of organizations, I've gotten innovative about these common challenges. With any new technology in this challenging environment, who's out there doing it and really finding ways of unlocking, no matter the speed bumps or other barriers. David, your thoughts?

David Vallejo [00:35:49]:

Look, I think organizations have to be ultimately ready. Overcoming the resistance to change is one aspect, but the other aspect is data quality and having the data foundation there that AI can eat into. It's like a hungry monster, and it's getting very cranky if it's starving or it produces wrong results. So that is definitely one obstacle that can be overcome and should be overcome first, actually, and then picking the right use cases. Right. I mean, we at SAP established this business AI framework of relevant, reliable, and responsible, and these are very key pillars. I laughed about it because it's three R's and it's leaning into the r three. But it is very important, right.

David Vallejo [00:36:36]:

If you think it all the way through, experimentation is great, but think it to the end all the way. And responsible is also a key aspect. And a lot of reservations that people have is what about introducing bias into the equation? What about data safety and security? Do you get organization to pump out all that data into AI? That perhaps is something that includes personal information, taking really a serious step of if we deploy this in production for everyday use. Have we considered all these things? For us at SAP, we take that very seriously. We have a whole ethics council around AI, for example, and sometimes that means certain dimensions of the business. We're not allowing AI into this frontier yet. It would be great if you have applications and have AI scanning all the resumes and giving you like the best candidates. Well, but there is a reservation.

David Vallejo [00:37:41]:

Well, will that always be the best candidates or will that have a bias on maybe how we hired in the past, but not how we should hire in the future? Taking careful but yet optimistic steps forward, vetting some of the use cases, making sure that the data is there, that security has been sort of observed, and with that, you can drive with more confidence on this path of innovation. And I think that is key. For all the naysayers they say, well, what about this? What about that that you say, I have checked that. I have checked that off the list regulations, I think will be volatile for some time. So you have, for example, chat, GPT, the meta AI in WhatsApp in the United States. You don't have that. Actually in Europe they don't allow that yet. So there are regulations also that need to be kept in mind.

David Vallejo [00:38:38]:

SAP, of course, we're at the forefront in some cases. We're sort of forming and defining that with the authorities. So these are all considerations that also make companies choose the right partner. So you can work, of course, with a wild and crazy startup to do some experimentation. But is that now a capability I want to weave into my entire enterprise for my HR processes? Well, maybe think about that twice. Right? So that's why IBM as well as SAP, with all the dimensions that it takes to take that innovation forward, I consider a very solid partner on that innovation front.

Scott W. Luton [00:39:20]:

Excellent point, David. Okay, quick comment there, Ronny, and then we're going to keep driving here in just a second for the sake of time. Ronny, what would you add to that response?

Ronny Horvath [00:39:29]:

David shared, most challenges we see right now is probably expectation and understanding. And the AI journey begins usually from a top level, right? One sea level goes golfing meets the other sea levels.

Scott W. Luton [00:39:41]:

Ok?

Ronny Horvath [00:39:42]:

I have AI now. It does basically everything for me. So that's a misunderstanding. Number one. There's this one, this group that says, well, AI will solve everything and we're going to have huge cost savings along the way. And there's this other group of executives that have anxiety because they think we're building Skynet. And to find a middle way between both and explain them that, yes, AI can save you money, but you have to have a strategy that works out. And then also telling them you're not going to fight robots, okay? It's not there yet.

Ronny Horvath [00:40:10]:

Okay? So getting that middle idea of what AI can bring, this is what we see is the biggest challenge right now because it starts the AI journey and this is where the expectation lies. If a CIO or a CEO CFO thinks that AI will save them all of the money in the world, we have to be proactive about. And they're like, yeah, there will be a lot of savings, but this is what we can do at the moment and don't expect too much out of it, right?

Scott W. Luton [00:40:36]:

That's right, Kevin. I'm a save. I'm going to get this next question and I'm going to circle back to get your additional commentary here. I want to dive into data management for a second. It's so critical. I mean, obviously it goes out saying, what steps would you all recommend that organizations and supply chain professionals plan to take to ensure data readiness for AI powered supply chain solutions? David, I'm circle back to you here. What are your thoughts?

David Vallejo [00:41:01]:

Yeah, I mean, first of all, it's making sure that we have the right data, that it's tuned also for AI consumption. And I like what Ronny said, sometimes it requires some myth busting, right? So, for example, not necessarily. You have a relational database, and you put that into a generative AI. That's actually not really well tuned data. Generative AI cannot take the unrelated columns and numbers of data and put that into some logical sequence. It is not tuned for that. It is tuned for language and making sure that you have the right data to essentially have a purpose driven outcome. I give you an example.

David Vallejo [00:41:42]:

We've used our integrated business planning platform, which does supply chain planning and has a lot of the forecasting and capacity and a lot of data in there. We actually had a great idea is how about we take all the history, lock information, where the engine is talking. I'm looking at this customer, I'm looking at this location, I'm looking at this product, I'm looking at this inventory point. Think of it like a really big log file, and put that into the large language model. And now using that as a prompt into the engine, now you can ask very intelligent questions like, why is my customer service level so low? Right? So making sure that the data and the structure of the data is fit for purpose, for the specific AI, whether that is generative AI or the specific machine learning capability that it's well trained, and then being able to protect the data, making sure that the data contains the right level of information, that you make sure that there are some guardrails around what that data contains. Right? Is it maybe having personal information, credit card information, Social Security information? That may not be the best level of data that you want to pump into a system where then somebody can ask questions like, give me Kevin's Social Security number. I'm very interested in that. That I think these are some key considerations around data.

David Vallejo [00:43:05]:

And let's not forget about cybersecurity. When you're pumping out data into a cloud where you're not even sure where it's going, can you be 100% sure that cloud is not being prone to be hacked? I'm sure we all got these ladders where some organization or the other sends you the information. Unfortunately, our cloud got hacked, so making sure that, you know, the data is safe is probably for me a consideration that will become more and more relevant for people, you know, when they go beyond the wild and crazy stuff as they see cybersecurity becoming more and more a threat. Just the recent two weeks, you've seen some of these, and a lot of companies wake up to that when they had an incident. And it's just like, you know, do people have earthquake insurance? In California, where I live, a lot of them don't. But guess what? When the earth start to rattle, where we had some years, it rattled here quite a bit. People start going out and buy earthquake insurance, right? And the same thing with cybersecurity, you feel completely safe until the bank gives you that call and says, unfortunately, your account was hacked and your money's gone. Right.

David Vallejo [00:44:17]:

So I think the data safety and security is very important measure that people have to take.

Scott W. Luton [00:44:24]:

Excellent response there, David. You asked about Kevin's Social Security number. I've got it. I'll give it to you later.

Kevin Jackson [00:44:30]:

I'll stellate last couple of days. Everybody has it. I think I heard in the news, they said everyone in the United States, your Social Security number is gone. That's the importance of protecting your data.

Scott W. Luton [00:44:45]:

Well, Kevin, let's stick with you for a second. David laid out a lot there when it comes to data management. Your quick comment?

Kevin Jackson [00:44:51]:

Well, AI systems rely heavily on high quality, accurate and comprehensive data in supply chains and logistics. I mean, historically has been incomplete, right? It's been outdated or inconsistent across all the different systems that are used. So I would say, you know, your number one challenge, your number one job is data, data, data. If you don't have the data that's curated properly, this can lead to suboptimal AI performance and stupid insights, to be honest.

David Vallejo [00:45:32]:

And some of this is actually training, right? And we've started to use, for example, you know, chat GPT for marketing purposes. And it's eye opening, right? You can ask like, hey, can you write a blog about this? And it comes out like, boom. Well, now I have to just do a few touch ups. However, prompt engineering is interesting because there is a lot of training that still needs to be done for people to use the AI in a proper way to ask the questions in the right way. Right. If you ask an incomplete question, you get completely wrong answer. Or you ask a question to write something for a specific audience in a specific style, you add a lot of details to the prompt that gives a much higher quality answer. I think there is an evolution that future workers will learn prompt engineering, just like they learn spreadsheets today, right? It becomes sort of a all purpose capability that people need in whatever job they are to leverage these modern technologies in a proper way.

David Vallejo [00:46:38]:

So as much as the data needs to be there, I think the skills and harvesting the data using the right level of prompt engineering will be equally important.

Kevin Jackson [00:46:48]:

Yeah, and there's also a very important aspect of that. You have to know your AI, because many people don't understand that. Chat GPT. It's great, but the data is two years old. Much of the data is two years old. So if it's nothing, if you're trying to get access or leverage brand new data, then it won't be in that AI. And that's why maybe it's even more important to bring your relevant data, your current data, into your AI process. You can't just consume chat GPT to help today's supply chain.

Kevin Jackson [00:47:27]:

All right.

Scott W. Luton [00:47:27]:

A lot to think about with data management, as you would expect. Ronny, I know you got a. A burning thought to add.

Ronny Horvath [00:47:33]:

Yeah. I like what David said earlier about myth busting in that sense, because I think data is what we heard. And, Kevin, you said it three times. Data. Data. I counted it here. So we see a lot of clients that are right now struggling exactly with that point because they want to start with AI, but they know data is so important. So what they do is whatever they have pushed out for the last decade, cleaning up their data, they do it now because they want to do aih.

Ronny Horvath [00:48:01]:

But I think they're overthinking it in that sense, that they could have already started with a small use case in an AI study instead of cleaning up the whole house. Think about it that way. Smart houses, we know those technologies that you get where you can control everything from your iPhone and so on. Sounds so cool. But if you have a house that's built in 1980 or 1990, I would love to just tore down my house and build it new just because of making it smart. Right? But I don't think you need to do that. You can start with one room, start with it, lay new cables, do everything so that it's automated. The same thing applies to data.

Ronny Horvath [00:48:35]:

You don't have to clean up your whole company and come to a clean core. Start with something. Start with a use case. Define a business goal, and say, this is where I start with AI. And you will immediately gain back benefits, because all the other companies who now start the data cleansing process that they have pushed out for decades, they will get the benefits two years down the line. And that I think that is sad, to be honest. So that's what I would like to emphasize, that we need to buff that myth that there is too much data work involved. It is important.

Ronny Horvath [00:49:03]:

Absolutely. That does not prevent you from starting with the use case in AI.

Kevin Jackson [00:49:08]:

Thoughts all grow big.

Scott W. Luton [00:49:09]:

That's right. The selection process is critical. We've seen so many organizations get that wrong. Right? If y'all could both, and I'll start with you, David. Give me one quick tip. There's so many. It deserves its own show. But give me one piece of advice when it comes to selection, for folks out there evaluating and going through the selection process of AI powered supply chain.

David Vallejo [00:49:29]:

Solutions, well, I think the most important selection criteria is to clearly link it to a business objective, what you're doing right. You know, going past the experimentation space where you try to figure out what, what is even possible and play with it. But if you really want to invest into a use case in a more extensive way, that it's clearly linked to a business objective, whether that is productivity and that you quantify it, right. That you're like, I really want to invest into this because I see the promise and the outcome of the following, whether that is customer service, inventory, whatever it may be. So I think that is clearly something that we also see some organization missing in their strategy for selecting is like, what is the business objective, what you're trying to achieve, who owns it? And if that is not in place, that pilot is oftentimes sort of doomed to fail. Because then people figure, why? Why are we doing this, right?

Scott W. Luton [00:50:25]:

And that's where the burnout of the valuable human factor, they can't understand and answer that why? And get an answer that that makes sense to them, that helps guide what they do day in, day out. That's critical. Well said, David. Ronny, what would you add one piece of advice when it comes to selection?

Ronny Horvath [00:50:42]:

Start with clear objectives. That's what I would say. What I said earlier, the expectation management. Yes, AI can do anything. No, we're not going to build Skynet. But however, make up your mind, look at the use cases, prioritize and pick out one that you start with and see what value you gain out of it. And this comes back also to the data topic. As I said, of course it's better to have clean data.

Ronny Horvath [00:51:03]:

My personal wish is that all customers in the world will have a clean core. They will all be on Hana. They have only SAP solutions, no header system, landscape. Again, this is going to be so awesome because then it will be so easy to implement AI anywhere because AI Julia is on there, right? It's on Hana. It's reading, it's writing back to your database. You don't have any data warehouses in between any data lakes. This is where we, I think, want to be in the future. And I think David wants to be there as well.

Ronny Horvath [00:51:32]:

I'm not sure, I guess.

Scott W. Luton [00:51:36]:

All right. An emphatic yes at that. David says, okay, I'm going to ask all three of y'all as we start to wind down, because we have fast and furious finish here, folks. We've turned a corner. We're already in. We're well into the second half of the year. As we were talking earlier about the different seasons that I got wrong, 2025 is right around the corner. So I'm going to ask each of you all, David, Ronny, and Kevin, what does 2025 look like for digital supply chains and AI? And David, your thought?

David Vallejo [00:52:02]:

We're around the corner of putting in our digital assistant Joule into our supply chain applications, and then we're doing this at the end of this year. So next year you will see that it will be the new normal that you log into your application and you have a digital assistant that will guide you through your daily work and where it answers questions, how to do something just like it became normal. You know, when you call your airline, it offers you a chatbot. And that chatbot is not as dumb anymore. Like in the past, right? I mean, I. I used it with rebooking a flight, and it says, I pulled up your itinerary. Here's an alternative option. Do you want me to go ahead and book that? No human was involved.

David Vallejo [00:52:43]:

So we get more and more into the sort of autonomous supply chain where some of these processes will be not just guided by AI, but actually starting to get actually into the level of confidence to automate some of these activities. That will be the new normal. I think 2025 will be a key year where we get basically an adjustment in the way we're using software altogether.

Scott W. Luton [00:53:09]:

Yes, well said, David. Love your comments around chatbots. That's been my experience as well. Ronny, 2025 digital supply chains, AI. What does it look like?

Ronny Horvath [00:53:19]:

IBM does so much in the sense our whole research department is on the forefront with language models. I mean, we all talk about large language models. IBM just released in may their granite language models to open source. They're updated weekly, and I think they will be updated daily in the near future. So I think this is something that will further be developed in the next year, specifically those large language supply chain or large language models. They will change to specific models because do you really need a large language model that has, like, a key lime pie recipe to do supply chain work? No, you don't need that. You need a specific model. Right.

Ronny Horvath [00:53:59]:

And that's why I think small language models, if you can call that, will be the future as well. I mean, a language model that fits on, like, something you can port around, take around with you without having connectivity. Think about something in your ear to translate language simultaneously. You don't need to learn languages anymore. It's gonna be solved in the future. And I think this is the next step. And I don't even want to talk about quantum computing because that's going to definitely change supply chain as we know it. Think about all the planning problems that we had.

Ronny Horvath [00:54:27]:

The traveling salesman principle. Ten locations multiplied to millions of opportunities that you have there, or possibilities. This needs computing power, and quantum computing is already there. Combined with AI. I think 2025 will be that kid in this toy store feeling for all of us. It's going to be awesome.

Scott W. Luton [00:54:47]:

Ronny, I love it. And I would just add, global supply chain works better if key lime pie is involved. Just saying. All right, so Kevin, David, and Ronny shared some bold visions for 2025, which, again, is just a few months away. Exciting times. What do you see, Kevin?

Kevin Jackson [00:55:06]:

Well, I see the most important aspect of AI is how it actually blends into our everyday life. So AI bias is the most important aspect in that. And I think we're going to see a lot of focus on preventing AI from perpetuating biases that are present in the data that they're trained on. This can lead to wrong assumptions, wrong predictions, and unfair actions, and really lack of transparency. So I think that's going to be the most important aspect as we transition into this whole digital world, leveraging AI for our digital supply chains.

Scott W. Luton [00:55:55]:

Well said, Kevin. Eloquent, all three of you, all. Very poetic. Okay, we're going to do this. We're going to wrap and bid adieu to our friends David and Ronny. We're going to make sure folks connect in just a second. I've got a resource, and we'll get Kevin's patent and key takeaway on the other side, so stay with us just for a second here. But first, we want to thank both David Vallejo with SAP, Ronny Horvath with IBM Consulting.

Scott W. Luton [00:56:19]:

David, how can folks track you down and connect with you and keep the conversation going?

David Vallejo [00:56:23]:

Well, track me down on LinkedIn. Go to SAP.com dSC digital supply chain to learn about our latest innovations.

Scott W. Luton [00:56:32]:

It is just that easy. You'll be glad you did, too. Appreciate your time here. David and Ronny, how can folks track you down as well?

Ronny Horvath [00:56:40]:

Same LinkedIn. And feel free to message write an email. My email is openly available on LinkedIn. Happy to answer.

Scott W. Luton [00:56:47]:

Or maybe at an upcoming sun devil game. I've been watching that helmet behind you. It's fascinating. There, y'all. Check it out.

Ronny Horvath [00:56:52]:

We're playing the big twelve. We're gonna get really dangerous this time.

Kevin Jackson [00:56:56]:

Oh, wow.

Scott W. Luton [00:56:57]:

Well, hey, thank you to you both. We look forward to having you back. Really enjoyed y'all's perspective here today. David Vallejo again with SAP. Ronny Horvath with IBM consulting. Have a great rest of your day, man. Kevin, I had about 37 more things I wanted to ask David and Ronny.

Kevin Jackson [00:57:14]:

There was so much in that show.

Scott W. Luton [00:57:16]:

Seriously, we could have added electric cables to this conversation and powered six cities across the globe. All right, so let's do this. Y'all know now how to connect with Ronny and David. Y'all check it out. Check out the thought leadership. They drop across social and more. Kevin, we got a really neat resource we want to add and offer folks out there. You know we're all about resources, right? And this is a neat white paper from our friends at IBM entitled Generative AI.

Scott W. Luton [00:57:47]:

ERP study 2024, achieving enterprise scale improvements with SAP specific generative AI. Okay, Kevin, you're not getting out of your patented key takeaway. And it's tough to boil all that down to one key takeaway. What would it be? Kevin?

Kevin Jackson [00:58:08]:

Explore. Explore. Generative AI. Okay. And explore with focus. You have to understand what it can do and how it can help your business model, help your business processes, help you on your top line and bottom line. So the bottom line is just do it.

Scott W. Luton [00:58:28]:

I like it. I like it. I've heard that somewhere before. No, Kevin, I love that. And I also love. There's so much this conversation I thought was really practical. Right. You can tell David and Ronny are out there amongst helping organizations find success when it comes to AI and technology supply chain a whole bunch more.

Scott W. Luton [00:58:48]:

But that curiosity thing that we. We dwell on a lot on the front end, I think whether you're a professional out there, you know, fighting a good fight as supply chain pro, or if you're a business leader and you're managing teams and teams, I think finding ways to challenge yourself to be more curious or to instill more curiosity in your team, whether you're in supply chain or any other part of global business, I think that is a big part of successfully unlocking not just where we can target the right use cases, as we talked about, lot when it comes to AI, but many other. I mean, that's how we find solutions to all sorts of challenges, old and new. Right, Kevin?

Kevin Jackson [00:59:26]:

Yeah, absolutely. But you gotta keep focusing. You gotta keep working on those challenges. AI is one way of making it easier to deal with all these changes in our life, in our business.

Scott W. Luton [00:59:40]:

So if you like what you heard from the whole gang here, this whole conversation, especially Kevin L. Jackson, make sure you find a digital transformers wherever you get your podcast. And you can also find Kevin across social. Kevin, a pleasure to have you here.

Kevin Jackson [00:59:53]:

Thank you very much. This is. This is awesome.

Scott W. Luton [00:59:56]:

It was, wasn't it? It was. Folks, here's my challenge to you. Take one thing that we heard from Ronny or David or Kevin and put it in action. Your team's ready to change how business is done. Your team is ready to have more successful, consistent, positive, fulfilling, purposeful days. And it's our job to help empower them to do just that. So on behalf the entire team here at Supply Chain Now, Scott Luton, challenge. You do good, give forward, be the change that's needed, and we'll see you next time right back here on Supply Chain Now.

Scott W. Luton [01:00:25]:

Thanks, everybody.

Narrator [01:00:29]:

Thanks for being a part of our Supply Chain Now community. Check out all of our programming at SupplyChainNow.com and make sure you subscribe to Supply Chain Now anywhere you listen to podcasts. And follow us on Facebook, LinkedIn, Twitter, and Instagram. See you next time on Supply Chain Now.