GMT20260313-203546_Recording

Gerai, gerai. Šiandieną vietą su autonomasiškai podkastu turime Gerolde Jackson'ą ir, visur, mūsų CEO'ą, Claudia Browne'ą. Gerolde Jackson'as turi labai ilustruotas atsidūrėjimas ir atsidūrėjimas. Jis atsidūrė Oracle. Galbūt jūs atsidūrėjai atsidūrėjęs į to korporaciją. Į to mažą kompaniją, taip. Taip, gerai, jis atsidūrėja į ekonomiją. Ale tak, Gerolde jest w kosmicznym świecie i ma wiele wiedzy. Jesteśmy tutaj w kosmicznym świecie, żeby porozmawiać o AIs, prawdziwych AIs w kosmicznym świecie, o tym, co się dzieje w 2026 roku. Dobrze, że jesteś tutaj, Gerolde. Wiem, że Oracle, jak powiedziałeś, to jego własna ekonomia, tak? Możesz powiedzieć, że to sprzedaż, bez powiedzenia Oracle. Jesteś prezydentem produktu w ramach, nie tylko Oracle, ale dużych korporacji, które są w kosmicznym świecie. Może to miejsce, jak wiesz, z inwestycją, tak? Nasz cel jest w zasadzie pomóc twoim klientom dostać więcej wartości z ich inwestycji w twoje produkty, tak? Więc, jak to jest, jakby, dostać nimbo, dostać więcej inteligencji z tego, co już inwestowałeś, w przeciwnym razie, niż rzucać i zmieniać. Bo jest dużo firm, które rzucają i zmieniają, ale, wiesz, nasz fundamentalny tez jest taki, że są systemy rekordów na rynku sprzedaży, tak jak Oracle, tak jak AppDynamics, tak jak SAP, tak? I ludzie, którzy je używają, proszą o więcej inteligencji w tych inwestycjach. Więc, może, chciałabym usłyszeć od Ciebie, co widzisz teraz w tym rynku? To jest taki, jakby, szalony czas, kiedy wszyscy wstali i zaczęli nazywać AI. Więc, co myślisz? No, cześć, Pudziu, dziękuję. Cieszę się, że jestem tutaj. Cieszę się, że mogę z Tobą porozmawiać o tym bardzo czasowym temacie, tak? Widzę wiele rzeczy na rynku. I jeśli ktoś z Was, którzy słuchają, są studentami strategii produktu i rządzenia produktu, to wiecie o tym konceptie, że jest to domyślny patron designowy, tak? I, wiesz, to jest to miejsce, gdzie znajdziecie prawdziwe wzrosty, kiedy zostaniecie w połączeniu z tym domyślnym patronem designowym. I muszę powiedzieć, że jeszcze tego nie widziałem w tym konkretnym miejscu. Jest wiele firm, które próbują wyjaśnić, jak używam AI, żeby uruchomić lepszy biznes. I kilka rzeczy, które podzielę się, a potem wrócę do Was, które są, po pierwsze, dlaczego firmy pytają o to pytanie? I, w zasadzie, macie dwa lub trzy warunki. Jedno to warunek, w którym jestem naprawdę dobry w funkcjach sprzedaży, a moja strona jest na mnie, o strategii AI. Mówią, że, hej, nie wiem, co to jest, ale słyszę to na nowościach, i jeśli otrzymamy strategię AI, to pomoże nam w sprzedażu sprzedaży, to jest to, co powinniśmy robić, a jeśli nie otrzymamy tego, to będziemy późni. Nie możemy się doczekać, byśmy byli późni na tym. Więc w tym kierunku jest wiele nauki, eksploracji i eksperymentacji. I naprawdę, wiesz, firmy, i to są firmy w wielokrotnym rozmiarze rozmiarów, od firmy w wielokrotnym rozmiarze rozmiarów do małej firmy, próbują to wyjaśnić, ponieważ technologia się tak szybko zmienia. Więc to jedno. Druga rzecz... Chciałabym zapytać o jedną z Twoich pytań. Widzimy dużo tego, co jest takie, że, okej, potrzebuję strategii AI, ale, wiesz, Ty i ja wiemy, że ludzie nie kupują technologii dla technologii. Może coś możemy zrobić, prawda? Oni szukają rozwiązania problemu, prawda? AI jest znaczeniem do tego, prawda? Więc jak słyszysz o strategii AI, co to znaczy dla nich strategia AI? Dlaczego chcą w nią inwestować? Czy to tylko formalność, czy więcej niż to? Tak. I to jest zwykle pytanie, które pytałem, ale przechodziłem do tych ludzi, którzy są pod pressurą od stacji. I jest kilka rzeczy, które są naturalnym odpowiedzią na to pytanie. Świat jest po prostu chaotyczny. Rzeczy się zmieniają, rzeczy się ruszają szybko i sposób, jak to nazywam, rządzenia twojej składniki sprzedaży nie jest doprowadzający do przeżycia dzisiaj. Mówiąc o latach, to nie jest pierwszy raz, kiedy słyszyliśmy o AI. Jeśli wrócę do 10 lat, to już było na scenie. Ludzie już o tym mówili. W tamtych czasach to było bardziej o precyzyjnych algorytmach, a może o wielokrotnym urządzeniu, o IOT i tego rodzaju rzeczy. Ale teraz... W 2022 roku. Dokładnie. Ale teraz myślę, że ten problem rozwinięcia prowadzi ludzi do AI teraz. Kiedy słyszę ludzi, którzy chcą technologii, zawsze próbuję przejąć rozmowę i powiedzieć, co to za gra, w której jesteśmy teraz. Jakie są zmiany, jaki jest poziom dynamizmu? Myślę, że to słowo. Tak, to słowo. Jesteś w miejscu konsumenckim, gdzie się zastanawiasz, czy... Nie ma szans do Kateliny, ale zawsze używam tego przykładu, powiedzmy, że Katelina Clark przychodzi do twojej miasta. I musisz być gotowa do tej onslaughty. Ale powiedzmy, że zmienia kółko. Dwa tygodnie przed przybyciem do twojej miasta. Prawdopodobnie ona już nie przyjdzie do twojej miasta. Ona już nie gra. A więc teraz firmy zostają zniszczone, zostają zniszczone z problemami na Tik Tok, zostają zniszczone z rozwinięciami na bazie sprzedaży. Ale każda firma ma jakąś ekspozycję do jakiegoś poziomu rozwinięć. I widzę konwergencję wokół tej narratywy rozwinięć o tym, czemu oni muszą, czy jak oni widzą wartości z potencjalnego inwestowania w A.I. Więc zajmuje mi czas, by dojść do tego z wieloma firmami, ale w końcu chodzi o to, że potrzebujemy tej inteligencji i tego danych, tego krótszego danych latencji, żebyśmy mogli być bardziej odpowiedzialni w tym chaotycznym świecie. Wiesz, to ciekawe, to jest jak, wiesz, w softwarze, prawda? To jest jak, zawsze w systemach. Ludzie zaczęli, ludzie po pierwsze inwestowali w reaktywne. Prawda? To jest jak, jak zmniejszyć czas na reakcję do problemu, który widzę, który widzę, prawda? I wtedy, cokolwiek to jest, zaczęło się z alerty, zaczęło się z dashboardów, zaczęło się z, wiesz, pracowników pracy, i tak dalej, prawda? I potem zaczęło się zmniejszyć do czasu, który jest jak, nie, nie, nie powiedz mi, kiedy to się stało. Prawda? Nie chcę być pierwszym, kto to wie. Chcę, żebyś mi powiedział, zanim to się stało. Prawda? I tak, ponieważ czas, i rozwiązanie decyzji, prawda? To się dzieje w czasie, kiedy wiesz coś, które, zanim się to wydarzy, i różne drogi, które, które mogą być przejęte. Myślę, że to jest to, gdzie przychodzi inteligencja, tylko do Twojego punktu widzenia, prawda? To jest to, że AI jest bardzo dobra w tej linii decyzji, prawda? Gdzie możesz właściwie być proaktywnym o problemie, i wtedy nie tylko być proaktywnym, ale również sugerować alternatywne, ponieważ wszyscy wiemy, że być reaktywnym jest kosztem dla firmy. Tak. Jest, nie wiem nazwiska, i nie powiem preaktywnym, bo to brzmi trochę... Nie jest proaktywne, nie jest reaktywne. Tak, tak. Coś w między. Może smart, jak inteligencja. Tak, tak. Myślę, że tak. Tak, to jest to, to jest to, co chciałem powiedzieć, wiesz, my spędzamy dużo czasu z wyjaśnieniem, prawda? Jesteśmy w sprzedaży. Tak, wyjaśnienie. Prawda? Wyjaśnienie, tradycja. Mamy całą industrię zbudowana na wyjaśnieniu, w zależności od tego, czy patrzę z tyłu, prawda? Czy patrzę z tyłu, czy patrzę z tyłu, czy patrzę z tyłu, czy patrzę z tyłu, czy patrzę z tyłu, czy patrzę z tyłu, czy patrzę z tyłu, czy patrzę z tyłu, czy patrzę z tyłu, czy patrzę z tyłu, czy patrzę z tyłu, czy patrzę z tyłu, czy patrzę z tyłu, czy patrzę z tyłu, czy patrzę z tyłu, czy patrzę z tyłu, czy patrzę z tyłu, czy patrzę z tyłu, czy patrzę z tyłu, czy patrzę z tyłu, czy patrzę z tyłu, czy patrzę z tyłu, czy patrzę z tyłu, czy patrzę z tyłu, czy patrzę z tyłu, czy patrzę z tyłu, czy patrzę z tyłu, czy patrzę z tyłu, czy patrzę z tyłu, czy patrzę z tyłu, czy patrzę z tyłu, czy patrzę z tyłu, czy patrzę z tyłu, czy patrzę z tyłu, czy patrzę z tyłu, czy patrzę z tyłu, czy patrzę z tyłu, czy patrzę z tyłu, czy patrzę z tyłu, czy patrzę z tyłu, czy patrzę z tyłu, czy patrzę z tyłu, czy patrzę z tyłu, czy patrzę z tyłu, czy patrzę z tyłu, czy patrzę z tyłu, czy patrzę z tyłu, czy patrzę z tyłu, czy patrzę z tyłu, czy patrzę z tyłu, czy patrzę z tyłu, czy patrzę z tyłu, czy patrzę z tyłu, czy patrzę z tyłu, czy patrzę z tyłu, czy patrzę z tyłu, czy patrzę z tyłu, czy patrzę z tyłu, czy patrzę z tyłu, czy patrzę z tyłu, czy patrzę z tyłu, czy patrzę z tyłu, czy patrzę z tyłu, czy patrzę z tyłu, czy patrzę z tyłu, czy patrzę z tyłu, czy patrzę z tyłu, czy patrzę z tyłu, czy patrzę z tyłu, czy patrzę z tyłu, czy patrzę z tyłu, czy patrzę z tyłu, czy patrzę z tyłu, czy patrzę z tyłu, czy patrzę z tyłu, czy patrzę z tyłu, czy patrzę z tyłu, czy patrzę z tyłu, czy patrzę z tyłu, czy patrzę z tyłu, czy patrzę z tyłu, czy patrzę z tyłu, czy patrzę z tyłu, czy patrzę z tyłu, czy patrzę z tyłu, czy patrzę z tyłu, czy patrzę z tyłu, czy patrzę z tyłu, czy patrzę z tyłu, czy patrzę z tyłu, czy patrzę z tyłu, czy patrzę z tyłu, czy patrzę z tyłu, czy patrzę z tyłu, czy patrzę z tyłu, czy patrzę z tyłu, czy patrzę z tyłu, czy patrzę z tyłu, czy patrzę z tyłu, czy patrzę z tyłu, czy patrzę z tyłu, czy patrzę z tyłu, czy patrzę z tyłu, czy patrzę z tyłu, czy patrzę z tyłu, czy patrzę z tyłu, czy patrzę z tyłu, czy patrzę z tyłu, czy patrzę z tyłu, czy patrzę z tyłu, to the right. So you're able to see these tells, if you will, that let you know that something is imminent, and that gives you enough time to prepare for it and react for it, right? And the same thing exists, I think, in supply chain or in business, right? When there's news of a strike, right, immediately we can probably tell that that's going to have an impact to me on my inbound. Even if there's a rumor of it, I can start at least preparing, right? And so AI can help us be out there listening for the key words. And then, like, every morning, you can wake up, turn on whatever AI agent, whatever thing you got going on, and it'll tell you, these are the things to pay attention to in the world today, and this is how it may impact your supply chain. So it's a more proactive scenario planning sort of an engine that we didn't have possible, but it's now possible again. I mean, for the first time. I think that's a really good point, Gerald. You know, the trends that we originally were seeing was like, okay, like, how do you bring outside the walls, right, inside, right? Because, you know, there's like people know stuff's happening out there, right? Where humans are not very good at just being able to compute all of what's happening outside that is going to impact my work, my life, my supply chain today. So I think that's one aspect. We're very myopic. It's like, I have my business to do every day. Exactly. But then I think the interesting phenomena in my mind is like not just that, right, because that was a lot of like analysis, risk models, you know, there's an industry that sort of like looked at like, if I see what's going to happen in, you know, 30 days, how is that going to impact things? But then in addition to it, there's a lot of chaos, like, pardon my, you know, chaos monkeys even inside the walls, right, that you're not able to understand, right? And then all of that is happening, you know, back and forth emails. There's so much information. Decision making is still happening in emails and text messages and phone calls. And I'm like, how do systems like, you know, these systems of record deal with that? Because like by default, they're systems of record, which means the decisions are happening elsewhere. And then things are written back to this. So what sort of, what are the patterns and trends you're seeing there? Well, here's what I'm seeing. So on one hand, you know, I've got one foot in this research world, this academic, and then the one foot in reality. And so in the research world, the studies around resilience say there's two challenges with information flows that create littleness in your business, right? Like lack of resiliency. I'm saying both. One is information flow within a unit and information flow across the unit. Right? And so this goes back to the myopic conversation. Like I'm doing this. So let's just take global trade for a second. Right? And then I'll answer your question about how companies are using this, you know, technology to deal with this. You've got one group that sits in what I'll call global trade. Right? Their job is to keep track of tariffs and policy and stuff like that. And so when a new policy hits, they know that, you know, they're going to have to pay this much or pay that much or whatever. Right? So they know that. But what they don't know is they don't know the bill of material for all the finished goods that they make. They don't know what the actual exposure is to the business. And they don't know that if they made just a couple of changes to the product structure and do a couple of substitutions, they can completely avoid having to deal with that tariff impact. Right? Because they're buying different components, buying from a different supplier. And that conversation doesn't usually happen until a quarter later when they're looking at the financials. And then they go back to the product owner and say, hey, look, man, is there a way that you can help us, you know, reduce the margin on this? Right? And so there's this information latency that just naturally occurs when people work in their silos. So one of the most innovative things that I've seen is a company using an AI agent now that understands the, I'll just call it the physics around how products are made, that understands that every product has its HTC code classification, has a supplier relationship. So strategic sourcing folks know that they have alternative suppliers. And so somebody who's thinking strategically at a supply chain level says, we need to figure out how do we streamline the flow of understanding and decision making when something changes on the outside. So they sort of start building a workflow where they start building an understanding of all the dots that need to be connected. Now, they don't predetermine create a bunch of rules and ifs ands. Right. Because now you've got generative AI, you've got large language. Yeah, you've got MCP connectors and a brain that can reason across it. Boom. And so now they're building agents that look across functions. Right? When something happens from the outside or maybe the trigger is when the global trade person sort of flips a switch. Now you've got an indicator that goes to all the different parts of the supply chain simultaneously eliminating information disparity rule number one, or number two, which is across units. Right? And so they're starting to think more cross-functionally. They're using their understanding of their business and their business processes, their operating models, and they're using AI agents to really just connect the dots on critical information flows that they know that they're going to need to put in place. That's really interesting. I'll ask a follow-up. We've talked a lot about around the buying portfolio. The people who are doing the procurement, the people who are trying to manage their risk. Right? What are you seeing on the supplier side? What does their life look like? What really changes? How does AI meet them? There's the obvious one, which is like, okay, suppliers are buyers themselves, so okay. But then as a supplier, I can't imagine that as their business grows, they're being pinged in multiple ways, right? For updates, for knowing when things are. What is their relationship with AI? Yeah. I think the richness of the relationship with AI I think is gated with the strength of their true supply chain expertise. Right? Let me explain to you what I mean by that. I don't know, 20 years ago, a long time ago, we were introduced to this concept of mass customization. Right? Delayed differentiation, postponed management. These are like supply chain models and orchestration strategies. Right? That, you know, come through theory, optimize through practice, but like supply chain leaders that were really at the leading edge of their industries, whether it's high tech or industrial manufacturing, took that concept and then took a strong look at their business and said, how do we operationalize this theory so that we can return, improve your return on our investment capital? Right? So it's, you've got to really know supply chain and like end-to-end supply chain strategy and orchestration to get the most out of this. So this is what I'm seeing from some of the most, some of the deepest supply chain experts on the supply side. Number one, they need to find new markets for their products faster. Right? Let's say you're a manufacturer, you just got hit with a bunch of tariffs and all this other stuff. You want a new business. You need a new business in new markets that have less friction. Right? And so now you're using this on the same, similar but different, you're trying to, you're using AI to understand where your product may be able to tie in to where growth is in different regions of the world. So one, suppliers are starting to look a little bit more global. I know there's a lot of energy around local sourcing and near-shoring and that's usually from the consumer end. Right? They want stuff closer. But if you're in a low-cost region, right, you're a supplier in a low-cost region, that's valuable somewhere. Right? Where they can't make it, they still can't make it that less expensive. Right? So you're going to make connections. And so think about product lifecycle management, PLM, where you've got one bomb or one component that can be part of many finished goods. Right? And so that's one thing. They're looking for new markets, new opportunities there on the supplier side. The other is they're trying to say, hey, look, I'm really good at what I do. Let me manage, you know, sort of VMI. Right? Supplier-managed inventory. Now if the supplier happens to have the strength of process excellence and technology, they can say, hey, look, while I might cost more on the purchase price level, I can use technology to help reduce your working capital required to fulfill the needs that you have. So overall, return on invested capital equation is still palatable for your customers because you're taking on more of the burden and maybe more of the balance sheet risk because you're using AI to do a better job of managing inventories outside of your four walls. Right? So those are kind of the things. And what I continue to encourage folks is you're not going to find these solutions if you're outsourcing your AI strategies to your analytics team. Right? Your analytics team doesn't know about supply chain strategies. They don't know about it. It's got to be embedded in your workflow, in your decision-making. Because it's not an afterthought or a beforethought. It's all of it. No. Yeah. And usually, now I'm going to sound like an old man here, but you sort of got me going on this topic, which is I remember when I was a young Six Sigma Black Belt back in the old GE days. Oh, we all know the Six Sigma Black Belt. Right? And I remember some of the smartest statisticians that I'd ever met like would show up in a meeting and they knew all the things, all the ANOVAs and all the data collection methods and the gauge R&Rs. And we'd be like a plastic facility. Right? And we're trying to figure out how do we improve the quality of, you know, get the black specks out of the plastic. And the folks who know how to make plastic kind of very quickly could identify the top three things. But the data scientists who were the master black belts at the time, early data scientists, they couldn't tell you jack. They needed mountains and mountains and mountains of data just to come up with the top ten. To get to that intuition. And then they needed, you know, months and months and months of analysis to come up with some obvious things. And so the real power came when they started taking the people that really knew the business and gave them the analytic skills. They could go through and more rapidly identify the real problems and more rapidly identify how the business process needed to change so that they could make better decisions and go faster. We're in the same boat now with AI. We need to get these tools to the real experts in the business so that they can go through the rethink process. And I think probably 30% of the companies I talk to get this. Most of them are still outsourcing you to their analytics team. That's a really good point. Is that, like, how, are you still, your AI strategy means intelligence for the analytics team, right? So that's the, so maybe, you know, while we're talking about teams and people, right, is that, it's on everyone's mind, right? Is that when does, when AI is replacing jobs, right? Every day there's, like, you know, 30,000 people for data centers, right? I wonder which company that is, Gerald. But, you know, yesterday at, you know, a consulate, 1,600 jobs, this is everywhere, right? And, you know, on the supply chain side, right, it's like, I'm curious, right, is that on one end, you know, there's two ways to look at it. Is that, now you can do more, right? It's like, if your supply chain strategy was based off of just interacting with, you know, 10 suppliers, and now given how the macro climate and micro climate is changing, you now can interact with 100, 200 through AI. That's the positive side of it. The negative side of it is that, you know, it's the retirement clip, right? Is that, you know, how many of these skills are now agentified? And, you know, and how do you use that as IP for the company versus IP that walks away when somebody leaves? So, like, how are you selling? How are you thinking of that? How are people reacting to that message? Yeah, I call it the labor challenge. It's like a third of the big three things that companies are dealing with. One is disruption, geopolitical stuff. The other is just margin compression. Do the sort of like tariff impact. And the third is this thing around labor. And so, the best, it's tough to say the best, but what I like to see are companies that are thinking about the AI as an amplifier. We, you know, we have the job replacer and the amplifier. Those are the two sort of camps, right? And ultimately, the math is you don't need as many people to do tactical things when you have AI, right? So, ideally, you're talking about, well, how do I grow two, three X on the same employee base? And then the other question is how do I figure out how to reduce the learning curve of new employees and how do I retain the knowledge of the age employees that are retiring out without disrupting my business, sort of restating what you already said, right? And so, what I'm seeing is and I'll see what I'm seeing in high school is the same thing I'm seeing in business. I'm seeing folks that are using this AI to just get better, to get faster answers are really not amplifying, right? If they're using AI... They're at the edge. They're at the edge of AI. Exactly. They're using, like, super search, right? Mm-hmm. They're not going to realize a whole lot of value and if your job is largely just sort of look things up then that definitely puts you at risk if you're just sort of an information looker-upper, right? Like, that's a bad place to be, right? Information looker-upper. That's an interesting profession. Yes. I see what you're saying. Yeah. I like that, yeah. Over here. But the best students are the same as the best employees that are using AI and the best strategies for AI and that's where when your AI becomes a true amplifier to your research, to your options analysis, like, every job, there's three or four ways to do anything, right? Very few things are hyper-deterministic, right? Like, there's always, like, am I going to give this guy a discount on volume or am I going to do this? Like, what are my options? Right? And so people that are using AI to weigh options faster, right? To do research on, like, new customers faster, do research on, you know, product structures, like, repair methods faster. Just getting smarter faster is one of the ways that companies are actually leaning into this amplifier mode for AI. But this also requires you to have what I'll call, like, this sort of lean type culture where you know what standard work is and you know how to evaluate a procedure and you know how to look at somebody's day job and identify how much time they're spending looking things up and then how much time they're doing this and how many times did they make a mistake and then you're using AI to reduce the number of mistakes, to reduce the number of time wasted looking things up and now you've standardized that as a new way of working and that becomes core to your future state process. And so when you bring on new people, the same AI that's using you to do this research is also learning about how the work is done. So you can start asking the AI, hey, what do I do next year? Right, like, how do I do this? What is it, you know? And so new people don't have to have 30 years of experience but the more people you have using the same process, leveraging the same sort of AI infrastructure that has memory, you're going to be able to make your worst performer as effective as your best performer because they're using the same AI tool, right? So rethinking work process or work product or the ways of working with AI embedded with the eye of doing like process reengineering or process standardization, I think is the way that I've seen companies guard against the, that's how they prepare for the retirement cliff and they also showcase the power that's possible when you give this tool with some thought to your employees. Now let me say the anti-pattern real quick. The anti-pattern is I'm going to give GPT to everybody and I'm going to let every individual in my organization I'm going to democratize the use of generative AI so you know your job better than I do so you figure out how to use it. You figure out how to use it. So everybody is out there using their own personal large language models to do tasks at work and what you end up with is probably 50% of people are just wasting their time. And the ones that do figure it out, well they're on an island. You don't get organizational level scale out of that because you're not using sort of this whole basic standard work process reengineering mindset. You're just pushing tech out to everybody expecting them to figure it out. And I see that a lot and every time I see it I try to get ahead of it and help customers understand that there's real power in sort of like thinking about a process and integrating it into the process integrating AI into the process to get you where you want to go. Yeah. Well Gerald this has been very insightful. I mean I think we're all lucky to be in this age right now to be you know it's like going back to learning mode again, right? It's that okay I hear SAS again, I hear you know SAS is dead, I hear you know you think of a feature and before you know the feature is already obsolete. It's a different world to be existing in right now. Can I, I know we're probably out of time but can I just say something about the SASmageddon business? Please do. Shout out to all my investor friends out there that spend time looking at companies and prophesizing that because code can write code it's the end of SAS, it's the end of software. I just want to highlight a couple things. One most companies don't want to make software. Most companies don't want to make tools. Most companies just want to run their business, right? So like the fact that SAS is available so that they don't have to manage data centers and all that like that that's here. Cloud native software is here to stay. Now this is going to sound a little wonky, a little techno wonky but I think more people need to understand what does it mean to have cloud native software versus software hosted in the cloud. Right? So I do believe that software hosted in the cloud has a shelf life. Like that is not, 10 years from now very few customers or very few companies are going to be running their businesses off of software hosted in the cloud. What I mean by that is stuff that you used to run on-prem, you wrap it in a container and then you go host it on a hyperscaler. That's not a half-life tool. That is not going to be the future. Why is that? Well because you're basically, you've taken a piece of software, you wrap it up in a container and you host it. That software is stale, right? It's cracked. Data model that you had when you set it up is the same. The API is the same. Like everything is pretty much static, right? But now it's hosted. So at least you don't have to pay for the data set, right? Cloud native software is built in such a way where every layer is somewhat abstracted. So innovation can happen at the database layer and you can adopt it without opening up your UI, right? Innovation can happen at the UI layer and you can adopt it without having to change the database. But you've got so much optionality to embrace and adapt fast-moving technology curves when you're on cloud native software than you can with software hosted in the cloud, right? So SaaS is cloud native software. True SaaS is cloud native. And so every single SaaS company that's got good cloud native architecture, when you name the large language model company, whether it's Gemini, Lama, Anthropic, or whatever, and they're in an arms race to figure out who's better, who's going to get closer to artificial general intelligence. As those things get better, true SaaS software can rapidly adopt it. It's like swap it out. Go from VCT 2.2 to 5.2. Go to Lama. You can swap it out. It's like having an automatic update versus a low TNT. Exactly. And now your SaaS that you bought, like last year, is evergreen. It's ever fresh if you've got a legit proof. So let the large language models go keep doing their thing. Now all they're going to do is make SaaS faster and better, right? Because the one thing that SaaS providers have that, I'll just call it infrastructure providers don't, is the context of your business. And usually they have process expertise. So they're giving you solutions. So SaaS is not dead. I think we're going to see SaaS is going to get better faster. And the hosted software that's sitting on multiple clouds, that's where I think we're going to see a fairly rapid decline over the next five to ten years. Well, Gerald, this has been a fascinating conversation. Thank you so, so much for your time. I appreciate how you're always sort of the person who is so practical around the world, right? It's like there's so much data coming. There's so many decisions happening. There's a hype around AI. There's a hype around technology. But ultimately we're humans. We're buying stuff to solve problems. We're buying. And if you can keep going back to that as part of like your core basis, right? Then all the hype helps you navigate. You can navigate to this hype. So this has been so fun. Thank you so much, Gerald. Absolutely. Thank you for the opportunity to chat. It's always a pleasure. All right. Okay, guys. Until next time. Thank you again. All right. Can you stop recording, George? Okay. All right. You are awesome. You are so good, man. Thank you so, so much. Well, as next steps, what we'll do is I'll record my kids' stuff this weekend and then George and I patch it together. We will send it to you first if you would like it or we have a channel on Spotify. We'll just add it there and we'll send you the link. Yeah. I mean, I trust you. Just roll it. If you feel good. I'm not a good editor of audio when I'm producing, so get the cuss words out. I don't think I used any. No, this was nice. This was slow. It was so perfect. That's what I told, yeah. That's what I told George. I was like, just leave me and Gerald and we'll talk forever. So I think this was awesome. Hey, I am looking forward to when you're visiting in April. I'd love to see you when you're there, when you're here. I'll make it a point. Yeah, we'll do. We'll do. Our meeting is actually at Oracle Park. I'll be right in San Francisco. You'll be in San Francisco. So I'm sure let's, please put me as a priority and we will make sure we connect. I'll have to put my son Julian number one, but you'll be number two. After that. Family first. Understood. Alright. Okay Gerald. Thank you very much. Have a very, very good afternoon. Bye-bye. Thanks for giving me the best last meeting of the day I've had in a long time. Oh, that's lovely. Okay. Bye Gerald. Bye-bye.