 Welcome back to theCUBE's live coverage here in Boston, Massachusetts, Red Hat Summit 2023. Also, Ansible Fest folded in for the first time. I'm John Furrier, host of theCUBE with Paul Gillan, breaking down all the action. Day two, wall-to-wall coverage. CUBE alumni going back to 2015, an OG in theCUBE, Chris Wright, Chief Technology Officer and Senior Vice President of Global Engineering at Red Hat. Chris, great to see you. And thanks for coming. I know you're super busy. Coming off the keynote stage, everything's good. How are you doing? I'm pumped to be here. I always enjoy being with you guys, so doing well. So going back, I wanted to weave in the 2015 because there's so much that's happened. I mean, that's just a long time ago. Seems like just decades ago. But look at what Red Hat, you got Ansible Fest folding in that automation message. I mean, this is something that's been a mindset of Red Hat. Now it's in open source. Now part of the big tent event, scale, efficiency, all happening. We had automotive on earlier. Talk about latency management and edge. A lot of stuff going on. Give us the quick take on the action here. What's your favorite announcement? How would you describe this year's Red Hat Summit given all this action? Of course, AI hype. Yeah, yeah. Well, how would I describe it? Super exciting. I mean, it's really, really amazing to be here together with our community. It's developers, it's users, it's different businesses represented, partners. And part of it is just the energy being in the same place at the same time, which is fantastic. We have a lot going on. So we're talking about different ways that we can look at helping enterprises reduce some of their costs and really maintain, build a lot of consistency. Think about automation to reduce the cost of running the infrastructure and running the business. And then what can we do to help developers be more impactful and help businesses generally grow? AI is a part of that, but it's not the only piece. So a lot, a lot going on. We had Matt Hicks on twice yesterday. One, he came in off his keynote, but also I'm here with John Granger from IBM Services, and we had a customer on. Really amazing how that whole acquisition was all about the scaling up and also the access to more engineering, more integration. So much happening, event-driven, was a great announcement. That's kind of like the blocking and tackling and then light speed with the AI, which we saw at Ansible Fest last year with Wisdom, kind of coming to light. Is the speed of innovation faster now with the IBM relationship, more exposure, more action? How's the product acceleration coming for customers? What are the highlights? Well, we work across a broad partner ecosystem. And in this context with the Ansible Light Speed announcement, we're partnering directly with IBM as really the brains behind the AI models, the foundation models. But in general, I'd say innovation just keeps speeding up across the industry. We have so much collaboration happening in open-source communities and more and more people come into those communities and that just speeds up the pace of development. The way we're working together is focusing on helping bring tools that come from our entire partner ecosystem and the AI innovation that helps us focus on automation, make it easier to quickly automate tasks, is just a way to help build efficiency into the business. And by the way, I saw AWS, Ruby was on stage, we had Dell on stage, the appliance, a lot of action. A lot of announcements this week. Light speed, event-driven Ansible, the software supply chain is very interesting. What are customers saying? I mean, what seems, which of those seem to be exciting customers the most? Well, it's the, if we break it down into areas of, you know, there's the infrastructure piece, platform engineering, there's the developer side. How can we make developers more effective, more efficient, move quicker? In any of those cases, we see a lot of interest from our customers. And I think one of the questions they're asking is, we're living in a world where the economic climate is unclear and there's a talent shortage. How do we get the people that have the skills to do the work? And so making things easier and easier to consume as software technology can get more and more complicated. We're building these hybrid clouds, which are spanning multiple infrastructures, public clouds, on-premises, all the way out to the edge. So what we can do to build consistency at the platform layer and then create tools for either automation tools or developers to come in and use that platform and really quickly, you know, bring their business to life. That's the kind of conversations that we're having. And they're like, how can you help us move faster with the skill sets that we have? Obviously the AI was really popular. You mentioned light speed, but the data science open shift was there. What's the pitch for customers to get more visibility onto that product? What would you say to them to check out that data science open shift announcement? Yeah, well, open shift AI is a collection of capabilities that we have that really bring together data and applications onto a common platform. That's a Kubernetes platform. That's open shift. And you know, we've spent a lot of time building applications. We understand CI, CD pipelines. We know how to put applications into production. We need all of that same rigor and discipline in the enterprise to take data, train models, tune models, put models into production. And that kind of workflow is really, there's not the same maturity level in the enterprise. So open shift AI helps bring that platform capability so our customers can build that maturity and understand what it means to put a model into production, monitor the accuracy of the predictions of that model and then retrain as you get new data sets and as models drift from the dynamic world that we live within. As a CTO, of course, this is the year of AI and a lot of concerns out there though about transparency, auditability, reliability. As a CTO, how do you come at AI? How cautious are you when you come at integrating AI capabilities into your products? Well, we're pretty thoughtful about it. And you know, some of the things that you mentioned about the concerns, I think they're very valid. So we have to pay close attention. What we're focused on, especially in the work we're doing together with the IBM research team that's built Watson X and foundation models, it's taking specific areas and training models with curated data. So you have a much better view of what goes in and as a result, what comes out. And this is the difference between a general purpose generative AI tool that's trained on a broad corpus of data versus something that's really more domain specific. And I think that's a key. And so we're integrating more domain specific areas like Ansible YAML as a starting point. We're immediately moving to Kubernetes YAML for OpenShift and you can kind of imagine how we'll keep adding these capabilities into our portfolio, leveraging more of this domain specific view and even enabling our customers to take their own data and build their own models that are very within the enterprise focus. So I think this helps understand where the data comes, helps you feel more confident when you understand where the data come from that you're going to produce a meaningful result. And that training data has come up in our discussion, certainly in our reporting. If you have the wrong training data or you're too loose or too fast and loose with the data, you can contaminate the models. So you have to be careful with the quality of the data. Yeah, there's a whole discussion around poisoning data. And again, if you have a curation process and a governance around what data goes into models, you're stemming that concern at least in a meaningful way. And that's where we've been focused is building these more domain specific generative AI tools together with partners. With Red Hat OpenShift data science, you could moving the company in kind of a new area. Data science where it's not traditionally been focused. How are you convincing customers to give a look at that product? Well, there's a couple of things that I think really help bring customers' attention to the space. One is all the innovation that's happening in machine learning and AI. All the tools are built in open source communities. And because we have such a strong relationship with open source communities, customers are curious, what can we do to help them bring in those tools into their enterprise? So that's one piece. And then the other one is bringing together a common platform that supports data specific workloads as well as application workloads. And in the end, they run together. So you think about what we would call an intelligent application or business logic that's wrapped around an inference engine. The way you build and manage and lifecycle those things together I think is really important. And so our enterprise customers are looking at that as a need and we're able to show them you have the skills already to run this AI platform, it's OpenShift, and here's the additional tools that build on top that let your MLOps teams be more effective, we curate content, here's containers filled with notebooks and PyTorch and TensorFlow and all the tools that you would expect as a data scientist. You've had a long history with open source, being a kernel contributor and programmer. Matt Hicks yesterday, when I asked him about the AI wave, which we think is going to be the most significant of the waves of the three ways PC, web, and AI, he threw on the iPhone in the speech, I think that's still part of the web, but we can debate that later. He compared the AI movement to similar traction that the Linux kernel and the Linux movement had in the early days, it really went really fast. He thinks it can be faster. What's the state of open source today? Because if it accelerates as fast as we think it will, it's going to put pressure on the system. The governance, the people, the structure, it won't topple it over, but it'll certainly challenge the integrity of the stability of the system. What does open source have to do to make sure they're not driftwood, that they ride the wave properly? It's an interesting question. So open source is a way of doing development and the reason I say that is there is no the open source community. There's collections of developers that form their own communities and there's communities of communities and so it's a big distributed, resilient, kind of organic model. And as a result, growth is a fairly natural part of the open source kind of development model. The things that are, you know, certainly we'll have more people coming in to help contribute. We'll have projects that will have a lot of pressure and figure out how to- More code that's going to be shared by machines too. Well, I think that was exactly where I was going. Part of this is what we produce can help developers be more efficient at both producing code and reviewing code and even making recommendations. So here's a better way to refactor this algorithm into the existing code base that you have. So, you know, I think on the one hand, we'll create pressure. On the other hand, we grow through people and we're resilient through this sort of organic style of how open source development works. And then the tools that we're developing will help us then scale again. So it's sort of a, there's an inception. It should be efficient. It should be a human AI interaction. I think that's critical. We're not at a place where we're replacing the creativity of humans. This is about machine-assisted human intelligence. And I think of it as you and I have the exact same skill set. We have the same task in mind. You start with a well-drafted, rough draft. I start with a blank sheet of paper. You're going to move a lot faster than I am. And that's what we're doing with AWS. I was talking to Matt off camera about the Google Memo that leaked. Everyone was talking about that two weeks ago, a couple weeks ago, and they used the word moat. And when I hear moat, I'm like, hmmm, competitive advantage. Is there always competition in open source? And that's a good thing. You clarified why a moat and these ecosystems is not a bad thing for open source. Well, the way open source works is businesses put developers on projects. Within the project, we're collaborating together to make the project functional, succeed, have features that represent different business interests. So in the community, the focus is building the best technology. Different businesses then take that technology and decide how they go to market. What is their differentiation in that space? In some cases, you'll see the kind of open core model where there's something around the open source that's providing proprietary differentiation. What we do is 100% open, so we have a different take on that. And I think we all just approach the business opportunity from a different perspective. But the community is really the core innovation engine. And in the community context, that's about collaboration and building the best technology. Chris, I know you're really busy. You got a hard stop and your time is very valuable. Thank you for coming on theCUBE. Close us out and just give a quick teaser of what's coming down the pike in your mind from a vision standpoint. You got Project Kepler out there on the energy side. What's out on the horizon? Give us a quick sound bite on what's coming in. Well, a couple things we didn't touch on, security. So security is really critical and all of our customers are asking, what can we do to help them build trust in the software supply chain? So we're putting together tools that are in that CI CD space to help them understand where code's coming from and where it goes into production, which is really important from a customer perspective. And then the whole world is grappling with our global climate challenge and software and software systems, hardware at the bottom of the stack are part of that. So we're enabling power saving features as we always have with our hardware partners but then taking that further with a project like Kepler to give a systems level view of where power consumption goes so that we can help reduce the overall power consumption of applications and overall systems. So really important things that are top of mind from a board perspective, a customer level and it's all happening in open source communities. By the way, Etos was on theCUBE before you came on. They were really impressed by the visibility and transparency in the supply chain to your security point. That's a huge issue. Yeah, it's fundamental because you want to understand where code came from and where it's in production so that you have the chance to remediate issues as they pop up. Thank you for your time. Chris Wright here at theCUBE. CTO, head of engineering at Red Hat, Mr. Cube, Paul Gill, I'm John Furrier. We'll be right back with our next guest after this short break.