AI for JDE: Boost Accuracy & Efficiency with Document AI + Orchestrator

Transcript:
It could be a nominee detection. It could be more predictive analytics as I talked with the warehouse movements, [...0.5s] um, and the cost savings there. Um, but it's really exploring, you know, your data more, right?And again, it all goes back to having good master data management. That's the biggest thing, right? Is your data good enough to get trained? Our manual processes still slowing down your JD Edwards operations. What if AI could preemptively identify anomalies and streamline document management?Today on night your Grandpa's JD Edwards were joined by Drew Robb, AI Advisor at Erp Suite to explore how AI is revolutionizing enterprise paperwork from automated data extraction to real time anomaly detection.By the end of this episode, you'll understand how AI is already enhancing JD Edward systems to improve efficiency, reduce risk, accelerate insights, and [...0.6s] how you can begin this transformation. [...5.2s]Welcome to not your Grandpa's JD Edwards, the podcast where we simplify enterprise technology to bring you actual insights. I'm Nate Bushfield and today we are diving into how AI is transforming traditional document management and error detection in JD Edwards.Joining us is Drew Robb, [...0.5s] AI Advisor at ERP Suites who will share how AI is automating routine task, enhancing data accuracy and enabling proactive decision making in JD Edwards environments. Welcome, 1st off, how you doing today? Thanks for having me, Nate.I am doing fantastic today. Um you know, just working on a lot of presentations for customers and really showing you can't order the possible with these different AI services and technologies we're gonna talk about today, whether be document understanding, anomaly detection, machine learning, a lot of different ones we're gonna talk about today.Um, and also on a lighter note, you know, headed to Cancun here in two days, so really looking forward to that, some fun in the sun. So, yeah, thanks for having me a good day and excited and ready to talk about this, this cool new topic, right?So yeah I mean yeah, Cancun sounds like a lot of fun. I'm sure everyone is wishing they were in Cancun right now, so make sure to have fun down there.But you are our second reoccurring guest.Stew was our first, you know, not to first your bubble, but for those of you, for those of the people out there that haven't seen our first episode together. Can you share a bit about your background and what brought you into the world of JD Edwards?Absolutely. So, yeah, as they said, I'm Drew, Rob on the AI Advisor at ERP Suite. So I really help customers try to leverage artificial intelligence tools and technologies, but really be more of an educator and an advisor.And that's what you're gonna see in this podcast as well as other things we really do is it's all about advising, right? We don't just wanna build you a product a and AI tool and send you on your way.We wanna teach you how to use or promote usability, all that kind of fun great stuff, um, that you need. Um.So I've been at European Suites for about five years now. Uh, started around 2019, [...0.5s] um, started in the, actually the product space. And that's kind of where I got into more of the tools and technologies that we're working with today. Um, as well as that right after that, I worked in data analytic state consulting, um, for the company.And what kind of a cool thing we're gonna actually talk about today is how we level leverage AI services combined with data analytics, uh, to create what we coined as smarter analytics. Um.So we'll get in that topic to get today, so really just seeing how that has come full circle and really showing that kind of these data integration, data analytics tools that we've used in the past can be more leveraged and better utilized with these AI services that have come out today. So, [...0.6s] yeah, that's pretty much my background and what I've been doing.So, [...0.6s] yeah, looking forward to today in order to discuss, yeah yeah I mean, [...0.6s] we're glad that you're here. Obviously, [...0.5s] you're a subject matter expert in the world of AI, even more so than a lot of people at our company.So it's, it's honestly always great to look in your mind a little bit to see what's new and exciting and [...0.6s] the way that ERP suites, but also the rest of the JD Edward space is really growing and expanding into the world of AI. But [...0.8s] without further ado, let's get into it.What challenges do businesses face with manual document processing and air detection in JD Edwards?Yeah, a lot of times with that, [...0.4s] just to be Frank, it's a lot of our customers, and, you know, it's not all of them, but a lot of them are waggers in the AI space and really just the automation space in general.Um, and the biggest thing is [...0.4s] they usually do this, um, when they do document understanding or anomaly detection, a lavatory manual, right?We talk about Excel sheets, we talked about just integration or just inputting and manual, um, inputting into JD Edwards. It's all around automation or it's all around just manual entry.Um, and what comes with that is, is really the biggest, one of the bigger topics we're gonna talk about today when streamlining processes is is a, is a data quality, right, making sure your date is good up to date, [...0.5s] making sure it's correct.Um, one big thing that we talk about and what we really highlight as far as use cases is improving master data management, cause that's really, once you do that, um, your possibilities really endless, um, inside of JD Edwards.So that's definitely one place that we really look at. And and, and a lot of times, yeah, these companies are just doing, [...0.7s] you know sort of the, the [...0.5s] inputting data.And, and really the biggest thing that comes back to is [...1.2s] a lot of employees are just spending too much time doing these manual processes instead of doing more strategic [...0.4s] business things that they can do in their business.Um, and the biggest thing is we need to open up our employees to do this right, and use AI as a tool accompanying you, um, in your job, right.And I've talked in the past, um, and, kind of, another episode that's called an on AI risk and remedies, is about, you know, [...0.6s] not being scared about AI, taking your jobs, will realize it, and being a part in the partnership.Right, [...0.4s] um, [...0.5s] can actually help you throughout your day to day lives and your day to day job.So that, that's one of the bigger things. Um, with that, it's really again, it's just about automate processes, and we'll talk later on about, kind of, our phase approach when they boom AI. Um, but that's, kind of, the first step you wanna take, cause you don't just wanna jump into an AI journey and AI solution right away. Um you know, there's things you gotta really think about.And, and we'll definitely talk, talk about that here, um, shortly. Um, but that's the biggest thing is it all starts with that automation peach using these very cool tools that we're gonna talk about today [...1.0s] exactly and trying to get away from like the manual entry errors or even, I don't know, delayed anomaly [...0.8s] like detection or something like that. You know um.But at the end of the day, automation is taking over it. We did another episode with Mo talking about orchestrations and what that really brings to the fold. In terms of JD Edwards, um, which actually [...0.6s] came out, it was our second episode. So I mean, [...0.6s] we'll see how that all goes.But yes, it is, it is definitely a, something that we all care about in this space. We wanna take away the human air and a lot of this data entry stuff and just [...0.9s] try to manipulate it as much as we possibly can.That was, yeah, you brought up orchestration and I just want a real quick point out, like, with the power of orchestration is, is no talked about in that last episode with our AI services and combining those together.That's really the empowering piece right there, right? It starts with the automation that goes from there, right, um, into, [...0.6s] you know, autonomous CRP, eventually predictive analytics, you know, predictability machine learning, all those cool coin terms you hear about. It actually comes with [...0.5s] the, the mirroring with orchestration.So that's, that's great you talk to Mel because that's [...0.8s] a lot with his team, and, yeah, just, oh, yeah, power between the two is important.So, yeah, he's definitely an all known being when it comes to that stuff too. So he's been in space for a very long time, [...0.8s] but anyways, uh, why are traditional methods becoming insufficient in today's data intensive environments?Yeah, what we're really gonna talk about today which is really cool is, is is document understanding, um, and really what you think about in that space and is kind of those OCR, so optical recognition, right, [...1.1s] with those tools, really what they were is basically if you had a standard PDF or Excel file, word file, [...0.7s] um, that keeps getting Senate had to be standard. Um, you can only read one certain document with that, right? Um.And with this, it's really, it opens, it bridges that gap. And so let's say, you know, you can breathe, you know, expense report, it can tell the difference between expense report, a sales order, [...0.5s] you know, balance sheet, income statement. Um.It just really leverages that and you can train it on a lot of different models, um, to actually learn from historical input.So let's say, you know, a customers out there, um you know, sending documents in, um, and something maybe, you know, messed up, maybe they're missing a letter, maybe the price is a little bit wrong.Um you know, you're able to actually adjust with that, um, with kind of what we call vector databases, and actually, um, store the historical user, and put what they usually put in, and actually adjust it to what it should be.Um, and that's really the cool part about this, this is not just standard automation.It, it can take it a step further. Again, that's where we wanna start, and we talked about that before, um, but that's really, that's really the biggest thing, and it's really just [...0.6s] trying to make, you know, the complexity of inputting data and manual processes more agile.Right. And we again, we go back to automation cause that's what we start, but that's kind of the biggest thing there is, is really streamlining that, [...0.5s] um you know, and the traditional ones, again, as we talked about, it's all manual, it's OCR if [...0.5s] anything, but where we can take this further, and we can talk about it later on how we can leverage these, these technologies together, and, [...1.1s] you know, and make it, make it even better than just automation. Right?So I think that's the really cool part, [...0.7s] yeah, exactly. And obviously we're trying to get away from human air, we're trying to get away from that side of it.But also at the same time, we understand people that are still on the human side of it entering this [...0.7s] data in terms of [...0.5s] if it's a PDF, it's, if it's a contract, if it's specific things that are in there, [...0.7s] AI still accounting for that Humanaire still adjusting to make sure that 100%, [...0.6s] yeah, yeah, exactly.And that's, yeah, go ahead, I was just gonna say just real quick. It's, you know, it's adjusting for that human error. And, and again, it's not gonna be 100% correct right away. You got it, you got to start somewhere.And that's the ability to continuously train these models and still have that, as you mentioned, human in the loop throughout this whole training process who understands how the process works, whether it's just uploading a sales order or, you know, an expansion port, you mention contracts, people understand how that process works.They've been doing in their day to day lives for multiple hours in a day even, [...0.4s] right. So we wanna make sure that it's working correctly, and that's why we continuously train, but you still have to start somewhere.It's not gonna be 100% correct. And that's important that you brought the human in the loop aspect, cause we definitely talked about that ton in our last, our last podcast on AI risk of gravity. So [...1.0s] exactly, yeah, talking about training, and I think everybody is growing with AI.So if you haven't checked out that episode, please go watch it. We really dive deep into the human aspect and [...0.4s] how that an AI are really blossoming together, and how we can all grow together to reach a level of productivity that we haven't before.But back to this podcast, absolutely a little bit into AI role in modernizing JD Edwards. How does AI enhance document understanding within JD Edwards? Yeah, and this is, this is just the coolest part. So, so one thing and we have demos out on our website or we're going to have soon, right? Um.It's really just the ability to, you know, go vanilla with your JD Edwards, make everything standard, [...0.5s] right? It's getting rid of customizations and just adding simple buttons to upload a file.And, and the real quick, I just want to get in the technology behind that, how it out pretty much works. So there's just gonna be one button.Let's say, you know, you're operating with a P FORTY TWO TEN for sales order entry. You'll have the button on the JD Edward screen. Um, and with that, all we're gonna do is connect back to OCI services, other services we use now, there's AWS services, you can do the same.And again, it's using that power [...0.4s] of orchestrator to do the automation piece and connect back to those services and leverage that.So using document understanding AI services, we can connect back train those various different sales orders. And again, they can be different formats, right? They don't have to all be tabular.They can. There can be data anywhere and that's just the power of training, you know, training the documents. And then with that, you simply just inside a JDL, would you not even working inside of OCI or the cloud. You just clicking that button, you uploading files, right?You can upload multiple files, it could be one, it could be 3, 4, 5, depending on, you know, what a customer service or at might be working with. Um, you click that button, you upload it again using the power orchestration, it reads that document and seamlessly puts it into a JD Edwards table.Another really cool feature will do as well is once it puts the data inside the JD Edwards table, the P FORTY TWO TEN, you're actually able to see the attachment inside the JD Edwards, and it will actually attach [...0.5s] the document inside of there for more data validation.Again, that brings in the human aspect, especially when you're training it, [...0.6s] right?So you'll be able to see all the data making sure it's went in the right place, making sure especially it connects to, maybe you have, you know, pricing inside of JD Edwards making sure the pricing comes through for the certain items that are uploaded.And this is again, just one use case. We've been talking a lot with customers on as a sales order entry, and it's very simple, right? And again, all that power is inside JDL words, leveraging orchestration combined with the OCI services. And that's just really, that's the cool part, right? So it's very simple.It's when I demo, it's a five minute demo, it's the quickest demo you'll see just seamless, [...0.7s] seamless demo. So, yeah, I mean, it's crazy how on like how cool these tools have become and how easy it makes the process for, for someone actually knows every day.So [...1.3s] exactly, like, it's, it's one of those things that if you do it every day, you're like, oh, it's just remedial task, no big deal, exactly might be used to it [...0.6s] with this tool with a different things that AI is really bringing to the table.It can take, uh, seem like not exactly the easiest process and make it seamless, like you said, like it can just cut through all the BS and just go [...0.6s] from point a to point B in a matter of seconds, get out of your hands that way, you're not wasting time, you can do [...1.0s] better things with your time and be a better time manager just by utilizing some of these tools, which is fantastic.Um, but can you explain how [...0.5s] AI driven [...0.4s] anomaly detection works and its advantages over traditional methods?Yeah, I mean, this and this is another cool piece, so we talked about kind of the document understanding AI service that we use, an Oracle, [...0.6s] um, another cool piece that we, you know, another cool technology tool we use is called OCI Vision. Now this is pretty sweet, so this is where you, you take images.So let's say, you know, one, one use case we kind of worked on is detecting if police are sufficient or they have issues with them.Right? So something simple you can think of any product in your business, just taking pictures, images and upload them in the JD Edwards what this vision tool will do once you upload it. It will tell you if it's defective or not.So basically, if you go through, you can train on, I think we trained on like 5,000 or so images, um, you know, and, and basically said this one's good, this one's not, right.And again, it's, it could be less than, that, doesn't have to be 500. We just, we want to make sure it worked how we want to see how it work, we want to test the service. And again, [...0.8s] you know, we were able to get some great indications there and then start send more, you know, tens of thousands more images and, and then you can see how close it was.I think it was about 75% correct, somewhere close. Again, still the human, human in the loop aspect. I guess you said I'm just like, no, this, this is not, you know, this isn't a product that this product looks, uh, insufficient. This, this looks like a failure versus this product actually looks good.So it can start to recognize images, right, and it's really good for quality assurance, um, inside of your business, whether that be, [...0.5s] you know, in the manufacturing line, [...0.5s] um and, and, and other places, right, so inventory.So, yeah, I mean, it's, it's, it's a really cool tool um, another place we're looking at for anomaly detection.It's, kind of, using a different technology. It's a orca machine learning, [...0.7s] and that really goes into the anomalous transactions, the fraud, the duplicates, the, [...0.8s] you know, different financial numbers that may look astray, or maybe an outlier, [...0.6s] right. Credit may be off, discounts may be off, something along those lines.Again, I'm just, I'm just throwing out different use cases here. Um, [...0.4s] it's really just about, you know, looking inside [...0.5s] training on a mile, training, a dataset training, just a table inside JD I was, and start detecting those anomalies.I mean, with that taking a step further is being able to get rid of the sonomalies inside your data, [...0.4s] data. And again, we're not there yet, but it's somewhere where we're going.So, kind of, two different aspects, ones actually doing anomaly, yeah, images, and the other one is actually with your data inside of JD Edwards. So, [...0.5s] yeah, I mean, that's, that's pretty much where we're going, and again, it involves, still involves a lot of different training there.Um, one thing I really quick want to mention with the anomaly detection, especially with the Oracle machine learning and even, and even the vision aspect is where we're taking it with, with analytics, right, [...0.4s] being able to create a cafe one page inside JD Edwards to create visuals [...0.6s] inside of your, inside of your [...0.9s] JD Edwards environment, right and again, it's kind of like we talked about what the document understanding doing all the work inside JD Edwards, this is what we're gonna be doing too.Um, we're gonna create dashboards for [...0.4s] detecting sales anonymously, as I mentioned, or doing warehouse analytics detecting, [...0.5s] you know, if, if items either go to one warehouse that might provide more cost savings versus, not, versus a different one. Um, that stuff we're looking at that stuff.We're integrating directly in the JD Edwards as well. So [...0.6s] it's also using those standard JD, JD Edwards, you know, applications like we talked about but just making them better, not customizing, but enhancing them with AI is the biggest thing.Yeah, um, obviously, we were talking about tools and, like, loosely, but what tools really facilitate these AI capabilities and how do they integrate with JD Edwards absolutely. Yeah, and, and I, kind of highlight a few of them there, but it's kind of the same thing.We talked about the document understanding, so we're still gonna leverage when I talk about vision, it's kind of the same thing as document understanding.It's still a tool inside of, [...0.6s] um, OCI, right? But you're still gonna, it's all about still having orchestrator and that back end process, right, for the automation piece.And that's still the, you know, orchestrator still there uploading those product images inside J d O, which again creating, you know, creating that button inside JD Edwards, clicking that button, attaching product images up there and letting the vision service actually read through and detect if there's, [...0.4s] um, if the props are good or bad, basically.Right. So it's pretty simple. And then with the Oracle machine learning, as I was talking about with you more like your data, right, your fraud, um, you know, anonymous transaction sales, um you know, this sort of those use cases, um, that that's a little bit more, um, deeper.It's kind of like a phase 2 thing, um, which we'll talk about later how we kind of, what we've been talking to customers about quick rents, this one is more of, [...0.5s] we have a bunch of data.We want to run some anonymous detection, we want to do some productive analytics on it. Let's actually send that data to an autonomous state of warehouse. Right, just that data, right, cause we don't wanna send all your data there.Um, this is inside of OCI again, just a certain [...0.8s] subset of data. We're gonna train it using Oracle machine learning, um, with those various models inside of there.And then you from using Oracle machine learning from there, we actually connect it to a large language model, [...0.4s] um, inside of their tattoo, create dashboards for you, um, to go gather more insights about that data and getting it could be a nominee detection.It could be more predictive analytics as I talked with the warehouse movements, um, and the cost savings there. Um, but it's really exploring, you know, your data more, right?And again, it all goes back to having good master data management, that's the biggest thing, right? Is your data good enough to get trained, right?So that's, that's definitely a step we got to take before, and we can talk about that when we talk about how to really get started with these cool technologies. Um, but that's the biggest thing.It's, again, it's, it's leveraging orchestrations as well as OCI services, but still coming back to [...0.6s] the end user is gonna be working inside of JD Edwards, was there accustomed to.Now again, I mentioned JD Edwards, the engines can work there. These can also be leveraged in other places as well internal websites, right?Webex. I don't know why I would do that, but we're on Webex today, but I'm just saying, like, these are, these are capabilities don't just have to be in JD Edwards, and these solutions don't have to be there, they can be in other places as well.So, [...1.2s] yeah, [...0.5s] I mean, there's so much potential for where we're actually, where we are right now. Obviously, but [...0.8s] I'm excited to see what the future really holds.But obviously, we've covered [...0.7s] what the technology really does, but how should a business [...0.4s] think about, [...0.4s] like, actually rolling out this inside of [...0.7s] JD Edwards?Yeah, absolutely, and it's definitely, you know, talking to customers. It's all about starting small, right, and, and starting from, from the ground up, right. And, and it's definitely taking one single process.We talked about the sales order process a lot today, cause that's what is resonating a lot with customers. Like, you know, they're executives are wanting to implement AI to the business.When we're talking about the level manager, it's always like, oh, executives wants to implement AI, how can we get this going?Well, we have an AI journey that, that's a 9. 1 journey to define use cases. Well, [...0.8s] a lot of times people don't see that as relevant. They, they, they need something now. They need some quick return on investment.So it's really starting out and starting small crawl, walk [...0.5s] run is something that Oracle has been saying a ton. Um, so really starting that crawl and starting from that, from that ground zero, [...0.5s] I'll just, let's say, you know, we're gonna do the sales order entry. We're just gonna do that, [...0.5s] start with like 1,000 of them.I start with, you know, one or two different types of sales orders, [...0.4s] um, that might look a little bit different that we can train on, um, and start sending those in and again, have that human in the loop to make sure it's operating correctly.But then from there, right, so we can start there and that, and that's a good place to start.And even with the vision, you can do that as well, or even a subset of data as we talked about the predictive analytics, all these different, [...0.7s] all these different use cases are all these different technologies we talked about.Um, you can start in that very single spot of just starting with one single process, [...0.6s] and then from there, this can expand on the thing about a lot of the services they can integrate together. So let's say you start with document understanding you bring in the sales orders.Let's say a territory salesman, [...0.4s] uh, territory salesman manager wants to make inquiries on those orders. They wanna know what's happening out there. They wanna know what's coming in, [...0.5s] um, so they can talk to customers, maybe give their customers better discounts, um, or, or a better price rate. Um, but that's the biggest thing.We can connect those territory sales managers to a digital assistant patch. We help with that.I know we haven't talked about digital assistance. It's probably a later podcast, cause that's another thing we're, we're venturing into. But it's all about building out the, kind of, processes in leveraging these tools in the right place in the right way.Um, and the big, I mean, yeah, and, and the biggest thing is, and we'll go back to that. The biggest thing is the use case though the above everything else we wanna make sure, especially talking to customers, talking to, talking to people is, is the US case of the most important.It's important to have that discovery phase understanding what is AI versus not AI in this world, right, especially with the document understanding what's important within that world, what's important for the anomaly detection.Right, right, um, but again, you can start with automation, and then we can build from there.And faces is the biggest thing, so [...0.5s] that, that's probably the most important thing. Um, [...0.8s] and then, yeah, again, it all goes back to data. You know, you start with that prep work, clean your data, and then you're in a good spot to move forward as well. [...1.1s]Clean data [...0.7s] for all walk run. I don't know, man, I think, I think you might in another life be a great preacher. Um, because honestly, [...0.5s] the other day gotta start somewhere absolutely small start.You gotta be curious, you gotta go out there and try to make the little tiny change, and if it works, [...0.8s] you can expand the process, you can do so many things with AI, and especially with what ERP Sweets is doing right now.There's so much potential out in the world outside of ERP Sweets, like Chachi PT when it first came out, it could not do half of what it does right now, it is [...1.2s] changed over a billion times, I don't even know what one there are.I think it's technically chat GPT4, but it's, like four point [...0.7s] to point something that keeps going down the line of what it actually is.There's so many upgrades, there's so many changes you can make, but you have to start somewhere, [...0.7s] start small, build the big, you know, exactly, [...1.2s] but for in terms of ERP suites, we switch a little bit for this, [...0.7s] what are the [...0.5s] common, [...0.7s] like, phases or steps that an organization [...0.5s] can follow to do this successfully?Absolutely. And, and I kind of jumped into that little bit earlier and it's, it's really starting with that discovery phase, any repeat suites, we have a, uh, we have a Journey AI journey assessment to see where you guys really are at on your AI journey.Again, you can start from anywhere, you can start from ground one, you're just starting to learn about AI, maybe some AI 1:00, 0.1, um, you can be in the middle, you can, you can, you know, you can kind of know a little bit about AI education, but you don't have any maybe use cases yet. Um, or you can have a use case, you can have a roadmap, you can have a plan, right?It really depends, but this AI assessment will really assess where you're currently at so advisors like us can better assist you, um, when we work together to build out that AI journey for you. Um.The biggest thing that we're noticing a lot from customers though is a lot of this quick win, how do we get to that point? And that's really starting, again, you still need that discovery phase. You still need to plan out and build out a good AI use case.Again, we have to make sure that these use cases are good for AI, [...0.6s] right, and these document understanding ones are, are very good ones, and they're, and they're very easy to implement, they're very easy to automate, like, we've, like, we've talked about in the past.But again, it all comes back to that, you know, that kind of phase 2 what, what's important after you figure out a good use case?Well, you gotta make sure the data is good, and starting with these subsets of data, starting in that one single process [...0.4s] actually helps you understand that, it [...0.6s] helps you realize that your data is, you know, is good. And if not we can help you with that.It all starts with that master data management, that's a huge piece, but another one security, and we really didn't get along to security.We talked a lot more in the last podcast about that, [...0.4s] um, but it's really just understanding Eugenia with security, right, making sure that's, that's, that's good and, and, and down pat.Now the thing with the OCI security [...0.6s] and the OCI services, it is, it really provides and encryption as well um, so with that, I mean, when you're sending documents in from OCI [...0.6s] into JD, I was, it's all encrypted from, from that rest and in transit, so that's huge, right.So the security is there, and that's all about setting correct role based security, so the certain users only have access to those certain [...0.6s] applications, tables would have you.So security is another important piece, but again, it's still a quick win, cause you only focus on that subset of data for the data piece and the security piece.And with that from there, it's all about developing the solution. And again we talked about the, I, I talked through the demo, what, what it would look like.It's, it's very seamless. It's using the orchestrations and the OCI services combined into one, um, you know, and, and putting a button out onto JT Edwards.And that's, [...0.6s] that's pretty much, and that's the beauty of, and that's, that's what makes it streamlines that the development life cycle is, is a few months at post, so you can be, you can have a working solution in in two or three months again with training, still, still have that training in the same amount that, [...0.5s] you know, we still have to really think about.This is adoption as well, right, [...0.6s] you know, with these n business users we have to ensure that they're gonna adopt these solutions that we're talking about today and actually use them.Because they've been so prone and so used to doing their day to day task of the automated manual entries, [...0.5s] now getting a new solution, is this gonna scare them?Well, maybe a little bit, but they're not gonna under. They're now gonna understand how much of their job they're gonna be able to do now, now that they're not doing manual data transactions for eight hours a day or six hours a day, right?That might be extrapoling a little bit, but [...0.5s] that's the thing is they can make better decisions, strategic decisions. Now that, you know, we've started with this Ground Zero quick win automated task, so again, just everything is content, and you get that quick win into your business.And again, as I mentioned before with the digital system, there's other solutions out there. We can build off of this with other AI solutions, so, [...0.8s] but it's a good starting point.Yeah, so it's really about creating a controlled testable rollout, absolutely starting smart [...0.4s] learning fast and scaling when the time is right.Yeah, 100%, no 100%, yeah, yeah, when the time is right, when you're comfortable, right, everybody's comfortable, that's, that's another big pieces, [...0.5s] you know, everybody's really educated on how to use, you know, the Dai solution.Again, we're building the project product, but was it really deep? And, and, and how much is it gonna implement your business? And, and what's, you know, what's the, all right, we always come back to that, but [...0.6s] is it really gonna be efficient, is gonna be impactive, impactful?It's gonna be innovative for sure, and that's what's gonna get you, you know, ahead of your competitors. Um and that's the biggest reason why this bubbles AI, AI is going around so much.It's creating that competitive advantage with these quick wins because you have something that, you know, your other competitors might not have, [...0.7s] right? And then you can always just build off of that and build more. So [...1.2s] exactly scalable, I mean, it'll help, it can change your business.It really can 100%. That's what's, that's what's great about AI yes, it is a buzzword, it's something that everybody's talking about right now, but [...0.6s] the end of the day, it really isn't packful. It really can transform your business from a data standpoint.It can make all these remedial tasks [...0.6s] automate, automated and [...0.5s] really increase production, increase productivity. There's so many possibilities [...0.6s] for this one tool will call it, obviously [...0.6s] the one idea of AI is actually a multitude of tools.Yep. Um, but it really is game changing. And I understand why it is such a buzzword because there's so many, so many possible things that you can really do with it and utilize.Um, but if you're interested in leveraging AI to enhance your JD Edwards operations, download our free AI Starter Guide at learn dot E R P suites com [...0.5s] slash AI Starter Guide. This resource offers practical steps to begin integrating AI into your workflows effectively.That concludes this episode of not your Grandpa's JD Edwards. A big shout out to you drew for sharing insights on how AI is reshaping document management and anomaly detection and JD Edwards.But if you're ready to modernize your operations, download our AI starter guide. The link will be in the description below. Like, subscribe whatever you wanna do, but start your journey.The main thing that we want you to get out of this is start your journey. And remember, keep your tech modern, your day to clean, and your paperwork smart. Thank you. [...4.2s]