Bob Muglia, former CEO of Snowflake, discusses his book The Datapreneurs: The Promise of AI and the Creators Building Our Future.
Hosts: Carol Massar and Matt Miller. Producer: Paul Brennan.
You're listening to Bloomberg Business Week with Carol Messer and Bloomberg Quick Takes Tim Stenovik on Bloomberg Radio.
Well, there was a story on the Bloomberg Today was an analyst at with a note saying nothing is AI proof, but live sports may come close. And basically they're saying maybe sports as an asset and sports rights or something that maybe's protected from Ah.
Well, absolutely, I mean continuing to appreciate. One of the most read stories was about a Sydney, Australia based fund manager who said, I'm not going to buy you know, nvideo and the chip makers of the data center operators, but I'm going to buy stuff that can't be broken by AI right. And he said it's a very small field. There's there's few things, you know, but there are things out there, all right.
So we are so obsessed with AI right and how it's going to impact our world. So joining us to talk a little bit about it. From Menlo Park on Zoom is Bob Muglia. He's former CEO of Snowflake, the global infrastructure software company. He was there from twenty fourteen to twelve nine. Before that at Juniper Networks and then a Microsoft divisional president back in nine. He is a board director for a lot of companies, board memberships on Microsoft, Snowflake, Relation, AI and a few more. He's got a new book out. It's called The Data Datapreneurs, The Promise of AI and the Creators building our Future. And I apologize, Bob, I've had a hard time getting out. It's basically a play on data entrepreneurs. And I love the title. Welcome, Welcome, How are you good?
I'm good. It's glad to be here down the Bayer this week, so it's fun. I'm normally up in Seattle, but I'm down the Beyer this week, so that's great.
Are they talking a lot about AI in.
The Bay Area? That's all they talk about down here?
Well?
Rightfully?
So from your vantage point, should we all be talking about AI? And how should we be talking about it? And of course I'm talking about machine learning, generative AI, kind of the next level, if you will.
Well, I think it really has hit the vernacular of common people and it's become part of everyday lives with tools like jack GPT, and it's really just the beginning. I mean, this is going to affect us in a very deep way. I think we're going to see all sorts of wondrous things coming in the next couple of years. These assistants are copilots that are going to help us in a whole bunch of ways. Just about every application is going to be updated big, big time, lots of change in technology. It's the biggest thing I've ever seen in my career, actually.
So, I mean, it's absolutely huge, and the co pilots are one of the things I'm most excited about for Microsoft. Do you think there's a killer app that we haven't seen yet or are there just so many uses that it will be hard to pin one down.
Well, honestly, one of the killer apps is going to be searched, for sure, and you know we have an opportunity to see search truly reinvented with these answer bots that are being created. Bing was one of the first ones. There's a number of others out I have an investment in a small company called Perplexity that's in that space. But it's really the first time you've been able to take on Google, I'll say that for sure. I think the office is going to be tremendously updated. Really, just about every application is going to up be updated. I saw Salesforce made some major announcements a week or two ago. You're going to see just about every application updated. And you know company I ran for five years, Snowflake, is going to be making major announcements next week. Lots of stuff happening. So, yes, lots of apps. I don't know if it's any one app. I think it's many apps that'll be that'll be will be killer apps.
But Bob Bloomberg Business recovering in a story about Reddit, and basically it gets to a bigger question about who owns the data that is going to help make generative AI smarter? Right, that's crucial. What's your view on that.
Well, there's an awful lot of data out there, and one of the things that that people are learning about with these generative AIS and these these machine learning programs is the quality of the data really matters. So I think you're going to see corporate data becoming very important, proprietary data becoming important assets that people are using. You know, public forms like Reddit and and lots of public forms that are out there. Those are all that's data that people are using. They may try and monetize it in some ways. I think you see you'll see people starting to do that if people have sources of public data. But in general, people are going to be using their data as a way of generating proprietary information. And what we haven't seen yet is this really break into the enterprise. And that's a tremendous opportunity over the next couple of years.
Does this generative AI make different social media platforms more important than before? In other words, we were talking about kind of who's the top right place, right Matt in terms of social media? Does it kind of upend the hierarchy?
If you will? It probably will over time. I mean, it's hard to say. It's hard to say. As I said, you know, I can see for the first time, you know, applications like search which have the opportunity to be disrupted, and I think social media will be different because you're going to want to have AI summarize things for you. One of the challenges with social media there's so much garbage on it. It'd be nice to have a bot be able to say what's interesting in that that happened today, And so perhaps the media company that does the best job of that may wind up in the best position. But we'll have to see to find out.
I mean, Carol, how many emails do you get every day?
Too many? I don't even read them literally, And.
If I could have, I'm constantly typing eight nine go, eight nine go, which is spam and delete on the.
Bloomberg, I don't even I don't even delete anymore because it would take so much time.
It would be great if AI. That's one area in which I can't wait for AI. The one thing I bristlet a little bit, Bob, is when when Carol says make it smarter, I mean, it's not really getting smarter, right, that's a term that you use for a human and I think we.
But it is becoming right, it's the use of it's different.
It's the right word, the right word. Well, what do you do?
You share the concerns voiced by people like Musk and uh you know open Ai, sam Altman about you know, somehow this program becoming smart enough to subjugate humanity like that seems to me to be a ridiculous and b kind of takes the responsibility away from the companies that are running these programs.
Well, you know, we've as we've all grown up, we've been seen many dystopian science fiction things, Terminator being perhaps one of the most the most accomplished of all of those. I don't I don't know how you can you can be much worse than Terminator was to humanity. So there's a lot of there's a lot of of of literature and science fiction that goes into this. You know, in the short run, the real concerns are what people will do with AI. And people are going to do everything with AI, good, bad, and evil, and.
As they do with social media, right, like they do with.
Social media exactly, and and and AI can do some things faster than humans can, so there may be some new challenges, but it also can help to fight against those challenges too, And so you know, we'll see AI defending us against bad AI, good AI defending us against bad EI. In the short term, you know, the long term concerns about about existential issues about humanity are really about what will this intelligence get smarter than all of us and then become rogue and go against us. Now, honestly, I do think it's going to get smarter than all of us. It's on a path to basically be able to take the ideas and the intelligence of everybody in humanity and to put a lot of that together into a system that can be immediately trained and replicated and grow and things. So there is a lot of opportunity for this stuff to get smarter. But ultimately we will be creating whatever it is. And I've always been a believer that values will reign supreme in the end, and that while humanity does many things wrong, ultimately we come up. We never really screw it up that badly, and ultimately we'll make the right things happen. I think AI is going to do great things for all of us in the short and medium term and in the long term. Who knows what will happen. But I still have a very positive outlook. I'm definitely a techno optimist.
I love your optimism, Bob. We need more of that, Let's get more.
I also think it's it's important. Yeah, it's all about people. I mean, ultimately, the thing to realize is these these tools are being created by people, and we will do to it what we do with everything, And like I say, lots of great things and maybe some bad things too, but we'll we'll figure out how to control it and make it work.
I think about healthcare like the ability because healthcare is far from even as we know. It's an old story unfortunately, But I think about the possibility if you've got something and there's no way the medical community can keep up with all the research that's out there, and so the ability to somehow use AI to figure out, Wait, what's your you know, ailment and your the nuances to it? And then here's the reason. I don't know.
I just think it's so dangerous at first. Do you ever get a little bit sick of go on WebMD? Then like all of a sudden, you're dying.
But that's not as smart as it could be, right bub Yeah, web.
Md is Unfortunately, you go, you go on the web and you don't know, you know, you really know what you're looking for, and you see all sorts of nasty things. I'm dying.
I'm always dying, to be quite honest.
Talk about these co pilots. Think about it. Think about a physician's assistant copilot right that has access to all the latest medical research and can look at a scan, for example, and compare it to millions of other scans that have been done, and really, do you know, find things that people wouldn't be able to find? These co pilots are going to be incredibly helpful in medicine. It's one of the areas where we'll see some of the biggest advancements in the next five years.
I'm going to have a co pilot here in the studio, mat Me. As much as I love you, I'm going to have a co pilot, and IM just.
Going to tell you deserve one.
He won't be as you know, entertaining probably or whatever. Still with us is Bob Mugley. He's former CEO of Snowflake, also at Juniper Networks Microsoft. He's got a new book ass called The Datapreneurs, The Promise of AI and the Creators Building Our Future. Still with us on Zoom from Menlo Park, California. And I promise, I bet if we have AI to revenge, they can say this titles, so forgive me. I do think it's interesting though, about the data entrepreneur that are out there kind of figuring this out. Bob right now, and I feel like your vantage point and having worked with some visionaries very early on, likea Bill Gates or even a Sam Altman, who is now increasingly because of Open AI become a household name. Take us back there, though, to maybe give us some insight of what that era was like, as we feel like we are embarking on a whole new kind of data era.
Yeah. Well, the book The Datapreneurs is really about the entrepreneurs I've known that have made the data industry and today AI what it is. And the basic theory is that there's an arc of innovation that that has occurred really since the advent of digital computing, with continued new improvements, new technologies that have happened over time that have been advanced by these incredible people, and that has brought us to where we are today. And it has been this series of progressions. Back in the late eighties early nineteen nineties I worked very closely with Steve Steve Balmer and Bill Gates in the work on Windows, and back then there was a focus on something called information at your Fingertips. This was a vision Bill had come up with about how we would be able to find any information we wanted by just going to our PC and asking and running a query and asking a question if it really is. Back then, most people didn't even have email. Email was relatively new back then, the Internet had yet to be invented, and So this was a pretty novel concept. And while it didn't pan out the way Bill had anticipated it being, which was really a desktop centric, Windows centric view of the world, the Internet really changed that and really brought us information at your fingertips in every way we could have ever imagined. And now we have this incredible amount of information available to us, and in some senses, the largest problem we have is it's too much for us to digest and to really make sense of. And that's where AI can really help us, because it can go through a lot of that that data and turn it into information that is useful for us and and and and much much easier for us to consume. So it's been a it's been an ongoing trajectory. The world was very different back then. Uh, you know, I've often made the comment that I saw you know, I saw Microsoft from a from a bird's eye vantage point and got a chance to see it all. I've often said I saw the good, the bad, and the ugly at Microsoft. I was one of the twelve witnesses who was in front of Judge Jackson in the DJ case and David Boyce, and that certainly would qualify as the ugly in my opinion. But h and then I did penance on that for a couple of years afterwards, when when I kept going back to the Washington d C to clean up some of the last issues we had with protocol documentation on the DJ So I lived through all of that back then.
What is there going to be a penance, if you will, or a similar tough time when it comes to AI. Is there going to have to be a very significant role by regulators in your view?
Well, first of all, let me start by saying that that you know, the Microsoft case really created a lot of law and and and I think all the industry as a whole has learned a tremendous amount from the mistakes we made way back when, so they can avoid a lot of the challenges that we went through. You know, in terms of regulations for AI, there certainly will be some I think understanding what needs to be regulated is something that's still being considered. I think that there are areas where people taking AI and doing things with it that you know that are that are that that like deep fakes as an example, there may need to be some new laws associated with that, because if I understand correctly the laws that we have today don't necessarily fully cover the idea of somebody creating a complete image, a complete like image of a person and then having that image saying things that are just not true and clearly that that can't be allowed and we need to make sure that things like that are not permitted. So the laws will get updated somewhat. But actually most of the existing laws can be applied because people are behind these things. And if you can, if you can take it, look.
That what happened with crypto right like it got met, it's gotten messy.
Crypto is messy to begin with.
Though it's true, that's true. That's but do you feel like you know what I'm saying that it feels like there was quite a lag in terms of regulatory oversight.
Well, there always is going to be a lagged. There's always going to be a laged because technology will move faster than the government can ever move. And so we you know, we discover you know, the Sam Bankman freeds that happened. You know, that's that is something that that perhaps could have been caught. But now I think people have a clean, a keen eye to look for things like that.
So I am curious. You're the kind of person that if I was going from New York to the West Coast, I'd love to be sitting on a plane with because I would pick your brain and probably drive you crazy. But having said that, what is when people know your background? What is the question that they ask you most? When it comes to our new world and our obsession. It feels like it almost euphoria when it comes to AI. What's the smart conversation too that we should be having?
Well, I think the interesting question is that I keep hearing is what are the new innovations that are going to come about? What are the areas where applications are going to be transformed, and how is the technology going to transform it. One of the things we've learned in the last few months, which frankly wasn't clear in December or January, is that that this AI is going to be available to just about everyone at effectively zero cost you. It'll be advertising based in some ways, and services will come out that are available to consumers as a whole. And I think that you're going to have these services available to people really in all socioeconomic brackets. That's one area where it was not obvious a few months ago, and now it does seem to be clear. We've seen an incredible explosion of development of new open source technology, so it's not just coming from the big boys. There was an initial thought that you had to have spent tens or hundreds of millions of dollars to build these these models. That is certainly still true. But what's happening is is that open source models are appearing that have a lot of that work already done that can then be customized. So the interesting question is where are these things are going to make a difference in the next few years. And I think you know you're going to see it in just about every application, and it's not just technology coming from Google and open AI and Microsoft that'll make it happen.
Love this, come back anytime and open invitation, no doubt about it. Really appreciate it and good luck with the book. Bob Muglia. He is the former CEO of Snowflake. As we said, the datapreneurs, the promise of AI and the creators building our future. It's a new book, it's out. Check it out, so relevant to what's going on right now. On zoom from Menlo Park, California. Bob again, thank you so much. You're listening to Bloomberg BusinessWeek, watching us as well. I'm Carol Masser along with Matt Miller, and this is Bloomberg Radio.
M