 Hello, welcome back everyone, live coverage here at VMware Explorer, 13th year with theCUBE here, formerly VMworld, theCUBE's been there since 2010 when Paul Morris was the CEO, I'm John Furrier, host of theCUBE, Dave Vellante, on the other side, my co-host Rob Streche, leading our CUBE research team, The Collective. Now, you can see a lot more research coming out of our team, theCUBE and the data, our data leg and our technology. And of course, we're more excited to have our friend on theCUBE, Sanjay Poonan, CEO of Cohesity, formerly VMware, formerly SAP. You are on in the early days, 2010 with theCUBE and great friend, great to see you. Always a pleasure. Super excited to have you on theCUBE. As you know, open invitation anytime in Palo Alto, you're busy running Cohesity and hot market. How's it going? John, it's been great and congratulations to you and Dave and Rob on your success. Yes, 2010, I was at SAP and the name was still theCUBE, but I remember having a thoughtful, deliberate conversation. And you know what we talked about? The SAP workloads like ERP and so on, running on this cloud platform called AWS. That was our first conversation because AWS was starting. And then I remember Andy Jassy doing a little video piece for us because we put our mobility platform. And then in 2011, I think you were at AWS re-invent and it was the first time Andy had any external guest at re-invent and I was his first guest. So it's been great and yeah, the great fun at Cohesity. We could talk about that. It's been a year since I took the role. We have really cementing our brand as the best platform of the next generation of these data security management companies winning a lot of high profile customers. And as you know, you kind of build one success at a time. Well, before we get to the Cohesity story, what's going on there? I want to just comment on what you just said because if you think about at that time in 2011, 2012, 2013, cloud was not obvious. But for entrepreneurs and folks in the tech inner circles, yeah, it was obvious. Something's happening big here. Not like the AI moment now, but like not as big. It was poo pooed by everybody. I mean, at that time, I remember he said, oh, the cloud's a joke. No security in the cloud. I'm just talking to Raghu over there earlier. And he's like, oh yeah, I remember people were pissing on security all the time in the cloud and just how things can change so fast. And you've been involved in so many inflection points, Sanjeev. But I got to ask you because like, you've been on both sides of the conversation. Technology wave, business wave, bubble wave and managed through that effectively. What is your take on generative AI? Because it's a gift for many, if you're a tech company and you've been doing machine learning, old old AI as I call it, three years old, that last year. And you're in the cloud. This generative AI is a gift. Yes. What do you make of it? You've seen many waves. Yeah, I think it's exactly as you described. When in 2011, 2012, as I was at SAP, I think there was some denial of, how could this company called Amazon actually build a cloud? And of course the rest is history because what Andy Jassy has done and now Adam's continuing is nothing short of incredible. 85 billion dollar machine and Azure and Google have followed their footsteps in doing well too. We saw a continued wave since that time and prior in modernization of cybersecurity because threats especially ransom where we play very much in that wave. But I think generative AI, that third wave is going to be bigger than both of those. I think it'll be bigger than cloud and security because it's going to transform everything we do. It is the essence of being able to ask a computer to summarize the library of information and come back with something that's undisputable. So if you're writing essays and you wanted to summarize, I remember when we all wrote papers in college, we'd go back and pull out every book and even after the internet came back, we tried to scour everything. This summarization capability, the idea of summarizing a large amount of content which is the heart and soul of generative AI is so transformative of everything. And I've been playing around with chat, GPT and BARC. It's amazing what this capable can do and then finding the application of generative AI into enterprise software, like what we're doing now at Cohesity. We'll tell you a little bit what we're doing generative AI. I think this is going to be transformational. So I predict it's going to be bigger than what we've seen in cloud and cybersecurity. You asked me if I like to write before we came on camera and I'm like, you know, I don't really love it. But what AI does, and Dave Vellante quoted this problem we were on the super cloud event, he said it makes, because mainly the writing is the first use case for chat GPT, but AI makes a great person exceptional and a good person great. And so like I find that I'm a great note taker. I like to write, but I just don't like to script it up and put it into like, because I didn't get an English, I'm a computer science guy, right? So like, yeah, writing the perfect novels not on my bucket list item, but AI can help me. And so that's the use case for all aspects of foundational models. So cybersecurity. Okay, let's simulate an attack. Let's use AI to assist us. So you're seeing this augmentation, extension as the key value. How does that play in the tech app world? Because I can see the infrastructure, but the workloads are going to be AI workloads. What do they look like? That's the question everyone's asking. What does an AI look like? I think so that you're not AI washing, which is going to happen in any new areas. You're a cloud washing, you're sort of cybersecurity washing. I think I've encouraged CEOs of enterprise companies or bigger companies or small companies to really think through use cases that are very compelling. I think Satya Neda put it pretty well. We should think about general AI as a very good first draft summary of something. Okay, so if you're trying to say, summarize everything that Leo Tolstoy wrote and give it back to me, not war and peace, but in a half page, one page, it's going to do it better than anybody else. Or summarize all the works of Einstein and come back to me with a summary. So when you look at that summarization capability, about 12 months ago when this started to happen, I began to think that it had some application to Cohesity because in essence, you should think about a lot of the world's data protection backup data, secondary data, if it's on our platform, is sort of this tape that's very hard to search. So imagine trying to find that 1980s Michael Jackson song on a highly compressed data. That's what we've done. However, generative AI turns that on its head. Let me tell you how. I went up to Microsoft, spent some time with Satya Nader and as you know, I've been friends with them for many years and asked them, I have an idea here of how I think generative AI and open AI at that time could apply to Cohesity. What are your thoughts? In essence, if we have hundreds of petabytes or exabytes on our platform, highly compressed in our data, could we use open AI and generative AI to search that large backup of Cohesity data and come back with a summary to a customer of what they had stored in there for years? For example, let's say I'm a CISO of a bank and 10 years ago, there was a breach of my bank and the former CISO wrote up a whole bunch of documents about it, but it's in my backup. I want to use generative AI to search my entire backup and come back with a summary of what that CISO wrote. Beautiful use case. So here I am talking to something that's like, yeah, this is a use case in computer science called retrieval augmented generation, RAG. Go back and download all the computer science documents on that and implement it and you're on it. And I talked later to the AI team at Microsoft, later on to the AI team at Google. And in essence, that's what we've done. We've patented that idea because we think we're far ahead of everybody else's data protection and using RAG, retrieval augmented generation, to our industry. We're going to release it in the fall, we're super excited about it. And that application to us is going to be probably the biggest innovation Cohesity has worked out since Moet founded the company. It's remarkable. So that's how we've got to think. Now we're also doing some incredible things with AI and security. ransomware hunting, data entropy, detecting when you do a mass store, when you do more sort of surgical restores. All of those things are AI capabilities that we're building or building into the platform. So AI pervades a lot of, we do remember the roots of Cohesity are a bunch of Google engineers who started. So there's a lot of AI already in it. But this generative AI capability to us, I think is going to be mind blowing in terms of where we could take the company. And I'd encourage every CEO to do the same thing. I was going to say, and I mean, you guys, I was at AWS when you did the deal with them and landing on S3 and being able to back up to when store in Amazon. And I think when you look at what you're doing through the partnerships that you've built, the data is everywhere. And so how are you looking at it for keeping that data, especially when somebody's like backups, I was also, I ran all of the backups and DR for John Hancock, MainLife Financial in North America long time ago, I used to get letters from the SEC like on a weekly basis. Hey, I need this person's email from with the words market timing in these emails. Okay, now I got a phone. E-discovery. Yeah, yeah. So e-discovery type stuff. How do you look at it, keeping the AI, not only the results, but the data actually separated? Are you doing different models for you or looking to do different models for each customer? Yeah, really good question. I think the first way to think about it is, I think what happened in the last 10 years, every one of the traditional backup companies that had not modernized through a hyper-converged scale-out architecture was sort of left behind in architectural no man's land. And that's what Moet founded. We were the first to take his ideas of Nutanix hyper-converged to the secondary data space. Everybody else since that copied architecture. So we were the first to, so we have the best product today in an architected, cloud-ready architecture for doing data management. Now, AI adds on top of that the ability to do things like search or ransomware detection and security and so on and so forth. But we believe it has to be done in a responsible way. So if you don't have access or authority to be able to ask that query about what happened 10 years ago, you won't be allowed to do it. And that's what we call responsible AI. So you govern it by role-based access controls. You govern it with the appropriate governances. You only are allowed to do it on your data that you're backed up, not the world's data. So I don't believe that we're gonna be worried about the meaning of life questions, right? I mean, Googles have to not answer like is the computer sentient. We're not asking, you're asking AI-based questions on your data you're backed up and you won't be able to ask those questions. We are not allowed to do it because of security controls around it, right? So we're trying to govern this with a lot of controls. But even then, you're probably gonna have to worry about things that people get wrong. What if this summary that comes back to me isn't accurate, right? So you're gonna have to link to the source documents you found it from. So if someone says, listen, I actually wanna make sure that the summary that you found is accurate. You link, here's the actual document I summarized it from. You can read the detail behind it. You'll have a bibliography of where it found these sources from. And I think as we get better, the model will get better and better at being able to summarize with accuracy. I asked chat GPT when it first came out, summarize for me everything Sanjay Poon has talked about analytics. You remember my first foray at SAP was analytics. And it was, I came back with stuff that was pretty, you know, all over the map. I refined the search and it was actually not so bad, the second or third. And I think that's what's gonna happen with Genevieve in the consumer world. We're gonna get more and more tighter about refining what that person has spoken on a particular topic. I would love to say, hey, summarize what John Furrier has written in his entire or spoken about Andy Jassy. Because you've been one of the best at being able to interview him through his years. And that summary. I got a lot of content on it. Yeah, I mean, if you could summarize it, that's a good starting point to somebody else who's saying, I'd like to understand everything that Andy Jassy's talked about with John Furrier. If you could ask our cube AI any question, what would you ask it about you? What does Sanjay Poon in? What's the most controversial thing Sanjay Poon has said on the cube that got him in trouble? I don't know. With his PR team afterwards. I don't know. I mean, it's always that something. There's still time left in this. I know, we could go anywhere you want. I've always been reprimanded by my PR team. Like, you know, hey, you know, next time could you go easy on that topic? You know, you have to spice it up a little bit, but. See, it came up with a few things. It says accepting criticism is a thing you were big popular on making a decision during a layoff on someone. And then there was an interaction with D.R.A.S.H. Oh, let's talk about that. That's an actual response from our cube AI. D.R.A.S.H. and I finally, after all of that, we're actually, I mean, you know, I've not close friends with them, but I wouldn't consider anything. We were competitive. And when we finally met, we were at an advisory board. We're coming back on a flight. He said, let's take a picture and show to the whole world that there isn't this. And my view is in general, listen, we don't have to be competitively dismissive about all the companies. I think in my younger days, I probably did a lot of things, tweeted a lot of things that were stupid. And I'm willing to admit, like, whatever, seven, eight, 10 years older, I'm wiser about those things. I tell our people right now when we compete, to compete with integrity. I tell people among our competitors, they are good products, but now I'm going to tell you why I'm 10x better. But I don't have to dismiss them to talk about why I'm better. Anybody who throws fud out at companies and mud slings, they're not going to win long term. People know that you're a good competitor and you are actually in high integrity, so I can vouch for you there. I was just making the comment that we have our own language. We have 38,000 videos all with transcripts. We love it. And that's got a lot of charting in it. It's a treasure trove. So, you've got to be on it many times. So we think that's going to be augmenting our first. And isn't your son building the AI models now? No, he's pursuing the reputation. Oh, OK, got it. But I got some more four years inside the family that are doing all of this. I can take it everywhere. That's awesome. Really appreciate your time. I want to get a quick sound bite from you on your presence here at VMworld this year. And if you can comment on the state of the VMware ecosystem as it goes, this transition to Broadcom. Because the obvious trend is the bigger booths, a lot more bigger booths, and not a lot of mid-sized booths than a lot of little ones. It's almost like the mid-range of the ecosystem has been kind of hollowed out a little bit. So what's your presence here? I know you don't have to. Listen, I have a fantastic respect. Raghu came and spoke at our virtual conference. We did Catalyst virtual this time. I met for breakfast with Hawk Tan at a lot of respect, largely because Broadcom's big customer bars. They're one of our top customers. And Davidson, their CIO, they back up a lot of their workloads with us, so we're honored that they're top semiconductor company. So our view is that as VMware prepares for that, I think now it's a matter of when not, if it's likely to happen, they're getting all the approvals. I think that you're going to see, as he points out, there's going to be the public cloud players, and there's going to be the private cloud players. He's positioned very clearly that in the private cloud, VMware customers are in good hands. I believe that's going to be us. And our use case, as one of the top, we do VMs, databases, NAS files, and M365 better than anybody else. And our top workload is VMware. The largest customers who run on VMware back up and protect their data with us, we win them. Because our scale and speed are second to nobody, right? And then you add security like ransomware protection, threat hunting, cyber vaulting, we're ahead of people there. And now with AI, I think we'll take the next leap. So our mission, because we're much more an enterprise player, a lot of the other players are more mid-market in our competition. So our message to the top VMware customers is we're going to come here. We still think hopefully this conference continues going into next year. We will be of a presence here. We'll take care of our customers, whether it's on the show floor or on hotels nearby. I think coming back to Vegas is probably a good decision. I think there was sort of like, also Moscone, I think was kind of going through it's a little bit of, do you want to be there on that? A lot of crime there. Yeah, I mean, I didn't want to go there, but it's a tough situation. But I wish, I'm just saying, I only love for so many. I mean, I come here and I walk down the show floor. It's just so nice to see people whom I've known for many years, kind of taking pictures and selfies together again and high-fiving. So I feel like I'm forever part of the VMware family. And also the AI has given everyone a spirit of youth, youthful energy. I mean, Jensen was fantastic this morning, right? I mean, can I just say 30 seconds is the first time I met Jensen because in 2012, 2013 when I joined the company, we wanted to do a VDI integration with NVIDIA. And there were engineers at VMware who were not fully supportive. We got Jensen to come in 2013, I think it was a 2013 conference. He stood up on a table like this, guys, with his leather jacket, same leather jacket, he's wearing, okay? And just spoken, like 100 people gathered around him about the benefits of VDI passed through the GPU. And I was a Jensen convert from that point in time. You get hold court. I mean, I was like, he's a friend and mentor. I consider him, and it's so great to see company when I tracked at the time, his 20 billion market gap. When it was 200, I said, you're gonna be 500. And now he's almost a trillion. I'm very happy for him and that NVIDIA VMware partnership. So, you know, they're also a customer. We're gonna do more with them. So excited about this time. It's great to see you. Good to see you too. In person. Last minute we have left, put a quick plug in for what's going on at Cohesity. What's new? Give us some staffs. Give it a little commercial out for Cohesity. Yeah, I think, listen, we talked about this. We're getting more and more public ready. The big step is getting cash will positive preparing for when you're public companies about profitable growth. I strengthened the executive team. I made some announcement of a new CFO who was at great experience taking companies public. And, you know, we're winning in the market. I've enterprise proposition. Our customer base is phenomenal. We had Salesforce talking publicly why they picked us over alternative. So I'm excited about these new customers coming in. The way we win is we're a multi-cloud story. Okay, VMware is one, but then you've got AWS Azure and others. We are the best at security and AI. Those are the three vectors. A multi-cloud data management story. The best at cybersecurity for protection against ransomware. And now generative AI opens it all up. Those are the three vectors. You'll see us as we talk more as a potentially public company one day. And I look forward to the continued dialogue with you guys. Yeah, we're looking forward to seeing more of you. Sanjeev, thank you for coming on theCUBE. Really appreciate it. Okay, live coverage here. VMware Explore 2023 of theCUBE's 13th year, covering this conference formerly VMworld. We'll be right back with more. I'm John Furrier. Rob Stretchy, live day one coverage. We'll be right back.