In this episode, Bob Muglia, an Enterprise Builder and Author of The Datapreneurs: The Promise of AI and the Creators Building Our Future, answers every question you may have about the current and future state of generative AI. He talks about being a tech and humanity optimist, a Snowflake CEO, his new book with our very own Steve Hamm, and much more.
In this episode, Bob Muglia, an Enterprise Builder and Author of The Datapreneurs: The Promise of AI and the Creators Building Our Future, answers every question you may have about the current and future state of generative AI. He talks about being a tech and humanity optimist, a Snowflake CEO, his new book with our very own Steve Hamm, and much more.
---------
How you approach data will define what’s possible for your organization. Data engineers, data scientists, application developers, and a host of other data professionals who depend on the Snowflake Data Cloud continue to thrive thanks to a decade of technology breakthroughs. But that journey is only the beginning.
Attend Snowflake Summit 2023 in Las Vegas June 26-29 to learn how to access, build, and monetize data, tools, models, and applications in ways that were previously unimaginable. Enable seamless alignment and collaboration across these crucial functions in the Data Cloud to transform nearly every aspect of your organization.
Steve Hamm: [00:00:00] Welcome Bob to the podcast.
Bob Muglia: Hi, Steve. It's good to see you.
Steve Hamm: Yeah. I hardly see you anymore, you know.
Bob Muglia: Not as much as we used to see each other, that's for sure. Although, although we still see each other occasionally.
Steve Hamm: Yeah. Yeah, Bob. Bob and I worked together on his book, the Data Entrepreneurs. I helped him write it, and it's coming out like today. So yeah. So what have you been up to since you left Snowflake? What was it, four years ago?
Bob Muglia: Yeah, well, what I focused on is really working with a number of small companies that are in the data industry doing things that I thought were innovative and different. Uh, I really felt like, like with Snowflake, we had built something that really mattered to the industry. We solved a problem that had been incredibly.
Problematic for customers for many years, which is the ability to get all of your data in one place and have all of your users working on that data in a very consistent way, and have a consistent view of it, and really enabling companies to [00:01:00] become data driven. I mean, that was, that was the, the objective of what I was doing with Snowflake, and the Snowflake team has done a fantastic job of delivering on that and continuing to advance the products forward.
You know, I focused on problems that I thought were not solved. That were, were different areas. You know, an example of that is, is the space of complex data. Working with what people typically call unstructured data, images, videos, things like that. How can we actually get insights from those things? And, and now that's becoming possible with machine learning.
Steve Hamm: So are you an angel investor, a nurse? I mean, how do you, how does it work? How do you fit in?
Bob Muglia: Sometimes it's angel. I mean, that's early for me. I usually am a little bit when companies have, I can help become a founder who has a strong idea. And needs to productize that idea. They need to work, you know, figure out how to work with customers, how to bring the product to market, how they should price it, and, and, and, and who, how they would sell it.
I help, um, [00:02:00] entrepreneurs, CEOs that are, are, are working with very early stage companies, typically like what you might call Series A, although that those terms don't mean as much anymore. They used to have more meaning than they seem to today. Uh, but you know, not quite the earliest of stage where you might think of an angel, but where companies have a product that they're trying to get to, to be the minimum viable MVP product out, out in the market.
That's pretty much my sweet spot.
Steve Hamm: yeah, yeah. So you've been very busy throughout your career, you know, either running businesses or advising small businesses. You've never written a book before. So why did you decide to write your first ever book,
Bob Muglia: after I left Snowflake, I realized that there were a number of ideas I had and things that I had learned over the years that, uh, were useful for people to know. And I, I wanted to sort of get those ideas out there and. Originally I kind of thought I might write a few blogs or, you know, get a few blogs written that they're just targeted at, at different areas.
[00:03:00] And that's when you and I got together. This was almost two years ago when we first started talking and you know, if, if you recall, we were really loose in terms of how this might. Take what form this might take. It wasn't really, really clear at the beginning, and we really started a narrative and, you know, you were, you were, you were writing down what I was, you know, what I was, had learned over the,
Steve Hamm: Madly typing. Yeah. Yeah.
Bob Muglia: Yeah. And, and you, you started creating it and then you, you know, we saw together that there was a story here that could turn into a book and I think I had the idea of telling the story. What I wanted to do was, was, was. Help people to learn these, these things that I thought were important for people that are working with data and so write something that everyone would learn something from.
I think one thing about the book, the data is everyone who reads it will learn something of interest to them. And I think that's true. Whether you know you're Sam Altman or whether you are, you know, whether you, you you've, you've, you know very little [00:04:00] about data. I think there's things in there that every person will learn.
And, and I wanted to get those down, but I wanted to get them down in a way that that was fun and for, and enjoyable for people to read. And that's when we had the idea of telling the story from the, the, the, the, the concept of the data entrepreneurs I've worked with throughout my career. And that's where, where you came up with the term data entrepreneurs.
Steve Hamm: Yeah. Yeah. No, that was a, that was a good moment, a nice pivot point there. Yeah. I mean, for people listening, you know, the, the book is a bit of a history, but not a dry kind of history, but it kind of, kind of tells the history about the advances in. Data management and data analytics really starting almost from the beginning of computing, but it's, it's also partly a memoir because Bob, you were, you know, involved in many of those, those moments when, uh, of advance and, and some of the great ideas, some of the ideas that didn't work at various places.
And I think that's makes it even more engaging. And then of course, it, we bring it right up to the [00:05:00] present. Uh, you know, and it was just, I, I, I thought it was just fascinating the way it worked, that there we were kind of figuring out how to end the book and then late last year, all of a sudden, These large language models, these, these, uh, you know, these kind of major new advances in AI started popping and ever since it's been like a wildfire.
Um, so, you know, it's, AI is on everybody's lips as far as I can make out. We've had these huge advances. How do you address AI in the book?
Bob Muglia: Well, I tried to address it in the context of where we started from, which is telling. The story of these data entrepreneurs over a period of time that have led to where we are today. Cause it really is, there really is a history behind all these things. This didn't all happen in the last two years. I mean, it's been going on for decades really.
And the, it is a series of innovations that have occurred over. Really a 40, 50 year period that have led [00:06:00] us to where we are today. And, and, you know, book is driven by this idea that there is an arc of data innovation that has happened over time. And when I started working with you, Steve, on the book, you know, I thought about that arc from the very beginning.
But what I, I, where I sort of saw it, it ending was, was really the data economy and what we were building associated with. All of the things that data is doing to change the way people work and, and, and how it's enriching our lives and, and, and helping business in so many ways. And that's what I sort of thought the book would be about.
Um, But, but then, you know, what happened over the last year is this realization that that AI is advancing at a speed that is faster than certainly I understood it was, was advancing. I think most people have been taken somewhat by surprise in the speed of this advancement and, you know, come to realize how profound a set of changes this was going to have in people's lives.
Beyond where we were with data just a couple of years [00:07:00] ago, uh, with the potential advent for, for, for agi, artificial general intelligence, you know, machines that can think like an average person, that that is almost certainly going to happen in my lifetime, probably within the next 10 years. Uh, maybe even in less time than that potentially.
The, the profound impact of that meant that the book needed to evolve somewhat and, and, and ha and, and, and, and build that out. And so we, we, we spent a little bit of time, um, really talking about the future of what that, that can lead to, um, as we put it together.
Steve Hamm: Yeah, yeah, yeah. Now you're an optimist. You write that AI could help usher in what you call an era of Plenty for humanity. Uh, so what do you mean by that? Sketch it out for us.
Bob Muglia: Well, you know, I am an optimist. I've always been a technical optimist. I, I think that technology helped on. On the whole, it helps people to, to, to lead richer and more fulfilling lives. Mm-hmm. [00:08:00] Now, that's not to say every technical advance helps everybody in equal ways, cuz they doesn't, I mean, sometimes, sometimes these things, you know, cause harm to people in a variety of ways.
But overall technology is benefiting mankind. And, and you know, as an optimist, I was shaped a lot by the writings of Isaac Azoff, which I read considerable amount in my earlier days when I was in my. Teens and twenties. And he was also a technical techno optimist. And he had really envisioned a world, uh, of robotics, uh, devices, machines that are intelligent, machines that work for people and do things on behalf of people.
And in his writings, he talked about how. This technology could affect human society. And, and he saw it from all of the aspects of it, the complexities, the good and the bad, the challenges that it can bring as well. And so one of the things I sort of realized, if you, if you take the, the idea that we now have, Intelligence [00:09:00] that we can, we can package up in, in, in the form of a product in some form.
And intelligence has always been something that has been, been something only humans really had. I mean, we see it in the high order animals. I mean, I think you see it in whales and certainly you see it in some ways in dogs and things, but really true intelligence has been a human. Something that has been unique to humanity.
Now we see the beginning of the ability to, to make software do this. You know, that combined with the knowledge that we have, that we've built up over time and, and together with some of the other advances that could happen in, in reducing energy costs and then potentially using robots to, to, to lower the cost of labor.
It just means that, Everything will get less expensive. Um, in particular, you know, intelligence has, the cost of intelligence for some things has dropped considerably in the last six months with some of the artificial intelligence [00:10:00] products and services that are becoming available today. And that's just the beginning of this trend.
So all of these things have the potential to have tremendous benefits of society, but to society. But it's also important that we, we be very careful with these things.
Steve Hamm: Yeah. Well, before we get to that, and I do want to get to risks, you know, you talk about the era of, of Plenty For Humanity. You talk about the cost of these incredibly valuable resources going down. But you know, we've seen. A tremendous kind of split between the haves and the have nots.
The more and more of of the resources of the world, the money of the world, going to a relative few. Do you see that continuing in this, in this era of plenty or, or somehow? Does AI help spread it out?
Bob Muglia: Well, you know, when we, when when the chat g p t first came out, um, one of the first questions that I think people were asking was, was, is this technology going to be limited? To just a few companies. Is it going to be all in the hands of, [00:11:00] you know, a Microsoft in a Google and is it going to really be available to people across all walks of life?
And you know, while there's no question that there will be value that will accrue to the organizations and people that, that create these products and services in general, I think they can help everyone. And, and that the benefits will, will accrue across society. One thing that has happened is, is that, uh, the access to this technology is, is immediately became very broad.
It was immediately available essentially to anyone. And, uh, and, and really, I mean, there's, they're free options that, that exist for people to have, have access to this intelligence that didn't, didn't exist a year ago. And, and now anyone who really has a, a phone. All you need is a, is a smartphone can get access to it.
So I think it will be democratized. The other thing that's happened in the last few months that I'm very excited about is that, is that the open [00:12:00] source models are developing at a rapid rate. And so what what I now believe is that we're not going to just have. Google and Microsoft and Facebook, you know, meta, a few other companies that, that are the controllers of these large models.
I think we're gonna see thousands of models out there and that many different people will be able to create products and services and capabilities built on these different open source models. So I think we're gonna see, uh, thousands of flowers blooming and many people having op access to the technology.
we still have a world where, where resources are are not equally divided, but this is one tech resource that I think will be available to everyone and I think that's great. I.
Steve Hamm: Yeah, no, I, I'm, I'm glad to hear you say that now.
there are serious concerns about the risks put by AI and in your book and also in your, you know, when you're talked to people in, in all the formats, you have issued something of a call to [00:13:00] action.
How do you think the tech and business communities and society in general should address these concerns, these risks?
Bob Muglia: Yeah, I, I think, you know, we talk about, in the book, I talk about the need for new social contracts that, that. Exist between different forms of different parts of society, certainly including governments. As we start to think about regulating this industry, the, the, the, the, the thing I would distinguish here, and I think it's really important for people to, to, to separate these two things.
There is the use of AI as a tool used by people. For purposes that are, are whatever the objectives of the person are, and in that context, these are tools like any other tool that people have created. They're just very powerful in a variety of ways, and, and it's important that we recognize that that needs to be thought of as, as really an extension in [00:14:00] some senses of our existing laws and structures.
Because they're really meant with when people do, you know, when people use tools for a, for, for, for nefarious purposes to how do you deal with that? And, and like every other tool it's going to be used, AI is gonna be used for good purposes, for bad purposes, and in some cases for evil purposes by people.
And we need to regulate and control that. And here I can look and say that, that, you know, that Asma was way ahead of his time and he thought about these issues, you know, when he thought about his world. Which was filled with robots. Um, AOV invented the term robotics. This idea that robotics is, robotics is a technology that is created by people for the use of people.
That's what robotics is really all about. He really invented that term. And the three laws of robotics that AOV came up with, you know, in, um, in the early 1940s. Was, was the found is really a [00:15:00] strong foundation for us to think through how the, the tools that we create can work for the benefit of people, not against them.
The second thing, which is, which is which needs to be taken into account. Is a few years out. The, the, the things I'm just talking about now are very real today with, with the products and services, these co-pilots that are appearing. These are all very real, uh, tools and service products and services that will help people.
And also people will, you know, use for spamming purposes and
the deep fakes are going to exist. I mean, all this is gonna happen. And, and I think what will, what people will realize is that, is that we will just become better educated as a society.
Just like when Photoshop came out, you started, you used to be, when I was a kid, a picture was guaranteed. You had a picture. It was, it was for sure. And now when Photoshop, it was like, well, maybe not. And it'll just be a lot more of that right now. And certainly with video will be true. So that's the one case with [00:16:00] people.
I'm using it as a tool. Um, and they're the three laws can, can, can, can, can really help, um, for, for rutu. But the other thing is this idea that, that, and it's almost a certainty that these, these, these, uh uh, AI systems will continue to advance and get smarter and smarter. You know, it, eventually they will be about as smart as we are, and probably at some time after that, they will become smarter than we are and they will be able to do things that we can't do.
We see it already. I mean, they can do all sorts of things we can't do, but it's some, it's very fundamental because, you know, our neurons operated at one speed. And the circuits, the electronic circuits that, that these things are based on are thousands of times faster than, than the way our brains work. So it is not surprising that as this technology advances, it could have capabilities beyond what we have.
And here we need to think about how we [00:17:00] align these new entities. We have to think about them as independent entities that we've created. And to treat them with the respect that we treat other creatures on our earth and recognize that unlike other creatures, these ones are not, are potentially our peers or maybe someday our superiors in some ways and, and have capabilities beyond what we can do.
So we need to make sure we fully align them with our goals. And this is where. Asmo in his later writings. This didn't happen until the 1970s. Um, and when he, when he wrote his robot novels, um, that, that he, he came up with Zeroth law, which is this idea that, that, that, that robots. Cannot harm humanity or allow humanity to come to harm.
It's a much a higher order. It's a higher order, uh, way of thinking. The difference between a robot and dealing with a person and the issues that a person has versus the broader issues of humanity. And, and Asmo really [00:18:00] focused on all of the nuances and, and some of the challenges that associate as these machines become more intelligent.
And I do really believe that, that his laws, including that zeroth law, can be a foundational guide for. For how we build the rules and regulations going forward.
Steve Hamm: Yeah. Yeah. You know, it's, it's amazing. I mean, I ju you know, this is a case where leaders of industry are going to Congress and asking for regulations, which is not very
Bob Muglia: Not typical. Not
Steve Hamm: we, we know how remarkable this moment is in, in the history of our nation and the hi history of humanity really.
Bob Muglia: Partly they've learned too though. I mean, really, I mean, we learned, I learned this the hard way when I was at Microsoft, when we were right in the middle of the DOJ trial and I was in the center of that. And you know, that was a real lesson, I think for the industry. It was certainly a lesson for us, but it was a lesson that was broader to the industry.
And subsequently, of course, we've seen what's happened with social media. And some of the societal concerns that, that this has created. And now the [00:19:00] subsequent regulation that is, is, is being discussed certainly in Europe and, and in the United States.
I think that the, that leaders like Sam Altman and others in this industry, Sacha, you know, uh, Sundar, all these elite leaders have come to realize that they need to partner with the regulatory agencies as they're building this technology. So, while it's surprising that, that, that, that, that this is happening, I'm not shocked by it because I think the lessons have been learned along the way.
Steve Hamm: Yeah. Yeah. Now, early on in thinking through your book, you said that one of the important. The kind of foundational ideas was something called foundation models. And I think this goes to, you know, we all remember when Mark Andreason from Netscape, you know, famous VC, said, software is eating the world. And you know, people kind of play off on that.
And, and the, the idea here is, well, foundation models and models are basically eating software.
Bob Muglia: Yeah. Right. We say that explicitly in [00:20:00] the book, it's that models are
Steve Hamm: Yes. Yeah. So. What's, tell us about, I mean, talk broadly about models. Why are, why are models so important in general and why are foundation models or large language models so key to where we are in in the world today?
Bob Muglia: well, you know, I can argue that that. Computing is really, is, is really two things. In some senses, it's, it's actually people use com computers to communicate, and computers are used to solve problems for people.
And when we, when we solve problems for people, what we're doing essentially is creating a model in some sense or another. And people have been modeling for centuries, right? When the Romans. You know, when they, when they figured out the viaducts, which are pretty cool, if you've ever seen the water flowing that, that they did back thousands of years ago, they modeled that, they figured out what they had to do to, you know, I'm sure they built little, the little test things and they figured out what is, and, you know, we do that today and, and as we [00:21:00] build devices and things, certainly, you know, you don't create a, a modern device, you know, an aircraft, a car.
Anything major without going through and doing the engineering modeling associated with that. All of these things are put in, in, in the form of models and, and. When we think about business process, when we think about what we do in our daily lives, in some senses, there is a model we have in our head for what we're trying to accomplish.
And in business, we try and codify that in some sense of ways, and we've been doing it in different ways over time. I mean, models are written, are written down in, in, in documents, they're scribbled on, on napkins, they're on white. Boards, they're all over Slack communications. I mean, they're everywhere except not always in a, in a place or in a space where you can put things together and understand it all together in one place.
And what's [00:22:00] happening is that that is becoming progressively more important, this idea of centralizing the. Information, the metae that exists in an organization in a central place is becoming progressively more important. You know, we, we solve the problem, you know, with the modern data stack of centralizing the data.
The data is now can be for any company and, and many companies have achieved doing this with products like Snowflake. The data is now centralized in one system and is accessible to everyone, but what are they gonna do with it? How do they, should they think about that data? What are, you know, what is the, the, the business process, none of that is centralized currently.
It's all, all over the place. It's in code, it's in Slack conversations, it's everywhere. So what I think is happening right now is more and more of these things are being put in, in a centralized form in some sense. And sometimes these things will take the form of. in a sense these [00:23:00] models that are being created, the, are the, these new AI models that are created are, are taking some aspects of our business and are incorporating that into the intelligence of the ai.
We will also take, and I think over the next few years, Build new kinds of databases that allow you to take the knowledge of, of, of your organization, all of the attributes of the business process, how your sales funnel works, what are the important metrics. All of those things can be consolidated in one place.
And I think that will be called a knowledge graph. And I think we'll start to see that come out more and more of what we do in the world will be centralized in these models that that predict. How the real world is going to behave and can track on a minute, almost a minute by minute basis, the reality of that real world against the model.
Steve Hamm: Yeah. Yeah. You know, it's interesting when I look back on the 20th century, which I participated in half of. [00:24:00] You know, I think one of the big flaws, I mean, it looked like it looked like a, a, a value at the time. A good value was that we tried to solve problems by kind of isolating them and say, ah, yeah, these, these are the dimensions, these are the elements.
Let's just solve this thing Very, in a very focused sort of way. And what we lost sight of was the interconnectivity of things. And I, it seems like models. Really are allowing us to see things more systemically and to address them more systemically. We're, we're weaving all these things together. The human mind cannot do it.
Bob Muglia: All
Steve Hamm: That's why we, that's one of the reasons why we need, why we need machine
Bob Muglia: No, it's exactly right, Steve. These things are, are more complicated than we can, as, as human humans can understand. And in fact, one of the challenges I think, of putting these models together has been that we need these AI tools to help us to do it. Um, I, I, I've come to, I've come to convince that that except for relatively [00:25:00] isolated cases, People need these tools to assist them in, in putting the models together.
And it's very fortunate that the, that the transformer technologies have come out in the time, in the timeframe they had. They, the timing is perfect. I mean, they can solve problems. Uh, these new large language models can be applied to, uh, solving problems that a couple of years ago, there just wasn't solutions to.
Steve Hamm: Right, right, right. That's a beautiful thing. My hope is that climate change can be dealt with. In a, in a, in a much more kind of rational and comprehensive way using some of these models.
Bob Muglia: Well, I hope so too. I think that we can learn a lot from, you know, from, from the models effectively. I also hope that, that that technology will. I, I'm becoming, I, I am, I'm becoming a, an advocate that fusion may be a solution in, you know, in the next 10 or 15 years. And that would be a major, major thing. Um, if, if that can happen.
Uh, and, and there's never [00:26:00] been so much, there's more innovation happening in that space than has happened in my entire life. I mean, fusion is always 20, 30 years away, and maybe it's less now.
Steve Hamm: yeah. Let's hope so. Yeah. Like quantum, like quantum
Bob Muglia: That's another big one. When quantum happens, the world's gonna change and when the world's gonna change,
Steve Hamm: Yeah, so you write a lot about the modern data stack in the book, and clearly Snowflake is like right in the middle of that discussion. So how do you define the modern data stack and what is Snowflake's Data Cloud?
What, what role does it play in it? And I know, you know, things, things have evolved since you left Snowflake, but kind of give us that, that broad view.
Bob Muglia: Well, snowflake has really grown to be a full data platform with a full set of services and capabilities for people to work with their data. So it's grown from it, from its original roots of data warehouse to take on multiple workloads, um, in a complete platform. I mean, I, I think of the modern data stack as is a set, as a set of [00:27:00] software services that, that deliver data analytics, you know, first and foremost as a service that you purchase. That's the first and most important thing is that you don't run it yourself. You purchase this, um, from some third party vendor and those services.
Take advantage of the scalability characteristics of the cloud. So unlike previous solutions, you can handle effectively any amount of data or any amount of users. And then the last thing, I think that's the last key characteristic of, of the modern data stack is the data inside it. Is modeled for SQL databases and it leverages the technology, the SQL technology, and the, the, the incredible flexibility that a se a, a modern cloud data warehouse can provide.
And in some senses, I think of, like Snowflake is the slicer DERs of data. Um, they can take any amount of data you have and they can chop it up any which way you possibly want it chopped up and serve it to you, you know, on the platter [00:28:00] that you want it served to. You know, if you want to use Power BI or you want to use Tableau, or you want to use a machine learning model, you can, you can leverage that from that.
And, and what's happened is, is that, um, with some foundational. Services like Snowflake, an entire ecosystem has been built around this and we now have many, many vendors providing a wide variety of tools, tools to, to, you know, data pipeline tools, uh, tools that, that do data quality analysis, lots and lots of visualization tools, machine learning tools.
All of these tools are designed to work together in a consistent way. And what's been really fun to watch is that. Is that, you know, from the leadership that we did with Snowflake and, and and the, the, the incredible changes that company has brought to the industry, it's now really fun to see that there are are five platforms that are kind of building similar things, you know, snowflake and Databricks plus the three cloud vendors.
And, um, I still have a fan of Snowflake and still and still [00:29:00] believe that, that the Snowflake team is ahead in many ways. But, um, it's great to see, you know, it's great to see a number of different vendors offering solutions that have similar cap.
Steve Hamm: Now, you mentioned a couple minutes ago knowledge graphs. Uh, I think there's a lot of disagreement about or, or confusion about what that actually is.
Bob Muglia: let's.
Steve Hamm: Yeah, there's confusion and so you write a lot in the book about knowledge graphs, so tell us about knowledge graphs, but mainly how they can be used along with the modern data stack and machine learning technologies to improve data analytics.
Bob Muglia: Yeah, I mean, a knowledge graph is really a database that can be used that, that, that can model the complexity of, of the relationships that exist in a, in a business. If you think about what you do with sql, you model data in sql. But you don't really model your business and, and your data model and your business model are not fully aligned.
And in fact, that's why your business model is really written typically in [00:30:00] Python or Java Code, and it's mapped into sequel. The idea of a knowledge graph is that you have a database that can mo fully model. The, the, the, the logic of the business and, and the characteristics of the business. And you can actually incorporate that logic into the model.
And so it's actually can be executable directly. You can do things, you can directly take action from it. And it's, it's, you know, it, it is a very powerful concept. I mean, this idea of an executable model has been on some senses, a holy grail of software development for 30, 40 years. I remember some of the original modeling tools, you know, that go back all the way to the 1980s and, you know, they were always lacked the, the semantic capability to execute all of the.
All of the logic that you need. And now by using some new relational technology, um, and leveraging the, the foundation of, [00:31:00] of relational mathematics, it's possible to build models that can fully describe a, a business and, and, and, and you can leverage those to, to actually, actually make decisions.
Steve Hamm: Yeah. Yeah. Yeah. Another thing that you mentioned briefly before was, was this concept of the arc of data innovation. And, you know, it's a, it's a graph and, uh, though not a knowledge graph, um, but you know, it kind of picks up slowly and then it gets really steep at the end. Tell us about that. Tell us about the, the how you see, how and why we're seeing this incredible acceleration of technology progress.
Why is that happening right now?
Bob Muglia: I drew it as an arc. Because I feel like that's what it has been. It's been a continuous, it, it, it's continuously speeding up over my career and, you know, and when I look back, you know, to what I was doing in like the 1980s, you know, when I first joined business, I.
You know, the [00:32:00] speed at which things moved was very, very different than the speed at which it moves today. And I mean, we, you know, I, we still did interoffice memos for goodness sakes. You know, when I first, when I first entered the, the email didn't exist in my company when I joined, when I joined the industry, my first use of email was when I came to Microsoft in the, in 1988 and, uh, uh, And I think about the way we communicate, the way we work, and the speed at which we work.
What's happened is technology continues to and to advance human society close more and more closely connecting us and allowing us to share I ideas with greater and greater velocity. And, and it's this, this, it's, it's this. Human society is, is increases based on our ability to communicate between us, our ideas, and then leverage those ideas.
So you have a good idea if you, if you tell me, I can leverage that idea and, and, and tell somebody else, [00:33:00] and, and our ability to do that is just keep speeding up over time.
Steve Hamm: Yeah.
Bob Muglia: happening now is that, is that with the fact that we have access to all this data, We can, we can improve our decision making capability.
If we can fully analyze that data and reach conclusions, the, the, the appropriate conclusions from it, and these tools that are coming out, these new AI tools, the large language models will simply facilitate that, uh, and make that faster and, and easier for us to reach conclusions. It'll help us to get conclusions from our data more quickly, and as we, we can make decisions faster, we can implement things faster and it'll continue to speed up and, you know, it feels like it.
Just even this year with the changes that's happening in ai, it feels like things are going progressively faster and, and I felt that over my entire career, but it just continues to accelerate and, uh, I think we're still, we're really at just the beginning of that exponential [00:34:00] curve right now.
Steve Hamm: Yeah. Yeah. It's interesting. I mean, all that when you talk about connectivity, just it's so powerful. You think about ethernet, what that did, the internet, what that did, and then, and then putting the device and all that into, into somebody's hand. Yeah. These were, these were some of the big, uh, some of the major advances and they were really enabled by connectivity.
It's
Bob Muglia: GitHub. I mean, just GitHub. Just GitHub. Just GitHub. Just GitHub. Okay. I mean, think about that. Think about that. All the source code in one pla in one place that everyone can share. I mean, it's just a massive thing that's happened that didn't exist when I started in this. I mean, even 20 years ago, it didn't exist.
Steve Hamm: yeah.
you know, you're a technology optimist, and I think you call yourself a humanity optimist. What makes you so optimistic?
Bob Muglia: when I was, when I was a kid, when I was young. Um, it was the middle of the Cold War, right? It was, you know, in the 1960s. I went to school in, in the Cold War, you know, I ducked and [00:35:00] covered right underneath the, the desk. I remember that you talked about that.
You and I both did this and, uh, uh, Back then, the biggest concern was that nuclear weapons would wipe, wipe out humanity. It's still a concern. Okay. Don't get confused. It's still a concern. Maybe it's a bigger concern today than it's been in a long time even, but, but it's a, it's, it is a danger that we've learned to live with as a society and we've created mechanisms to manage it over time.
I don't think that anything we're creating has more negative potential. Than a nuclear bomb. I mean, it's hard to be worse than a nuclear bomb. And, and so what we just have to be sure of is that as we create these tools that have some negative potentials, that we also put in place, the mechanisms to manage and control that.
And, and I don't believe that I'm, I'm not, I'm not naive to believe that things will never go wrong. They will go wrong. Mistakes will happen. Um, But we will [00:36:00] learn from that as a society, and we will recover and we'll be stronger from it. I believe that we, that in general, society learns from its mistakes, although history does tend to repeat itself.
Steve Hamm: Yeah. Yeah,
Bob Muglia: so we keep getting, you know, in general the tr you know, in, in, in general the arc, you know, the arc of society is bends towards justice. I, I do believe that. I do believe that.
Steve Hamm: I hope so. I mean, I, I, I'm a believer in, you know, uh, see the, be a realist, but also be hopeful. So, and then, and then, and do, and do things to warrant. Hopefulness. So I think we agree on that. Um, so we typically end the podcast on a lighter note. And you know, you refer to the fact you and I have spent the last two years working together.
So what were the best and worst aspects of working with me?
Bob Muglia: Well, the best aspect was that you were an incredible help in, in getting the book done. And, and, and I I never could have done it [00:37:00] without you. I'll say that for sure. And, and it, it, it was, uh, um, you, you did a great job of, of. Helping to bring in the color of the book. I mean, you know, I knew all these people are people that I've worked with and known over the years, but you helped in talking to them to bring out some aspects of their lives that I think are, are, are great and, and never would've, never would've appeared without your, without what you did there.
And, and I think you made the book a lot more readable. Um, the, you know, the, sometimes I got frustrated because I wanted to get a point across and you were, and you were somewhat difficult to get. You, you were, you were a little bit, uh,
Steve Hamm: Would I talk? I would talk over you.
Bob Muglia: No, you just, you just didn't listen to me basically.
You just didn't listen to me. But it was really helpful because what I think would happen is if you sort of look at the way we did this, the technology points and, and were, you know, were things that I focused on adding in and it was not unusual where I would write something and you'd say, what the hell does this mean, Bob?
I don't understand this. And so what was really [00:38:00] helpful was that I think you made the book more understandable to a lot more people and I, and I very much appreciate that.
Steve Hamm: yeah. Hopefully that'll
Bob Muglia: You painted the tail though a few times. There were a few times you were painted the tail,
Steve Hamm: yeah, yeah.
So, uh, this has been wonderful. It's been great talking to you.
And I thought there were a lot of interesting. Sometimes complex, but very important points to make. I think the insight that I think, I hope people really go away with is this idea that you brought up near the top of when thinking about ai, we have to kind of grapple with the AI that we have now and hopefully use some of the infrastructure, legal infrastructure, regulatory infrastructure, governance, infrastructure that we have to kind of like quickly.
Expand to address it, to address the whole notion of, of people doing evil or, or, or nuisance with, with, with ai. But then down the road with agi, uh, artificial general [00:39:00] intelligence that we have to really think in a, in a, as as, A profound way as humans can think about this relationship between us and machines and making sure that, that they're working with us and for us and not against us.
So that is a, that's a, as a wonderful insight and I thank you for it. And I, and I hope a lot of people are listening or reading, so,
Bob Muglia: I think they are, A lot of people are, I mean, this is one thing that's been very, uh, positive is the uniform view that everyone seems to have. That, you know, these are, are tools with great potential, but also. You know, they could be used for ne for, for, for, for nefarious purposes. And we need to be very, very, very thoughtful about that.
And then in the long run, you know, we're creating independent entities and uh, and we need to make sure that those entities work with us, not against us. So,
Steve Hamm: Yeah, I like that.
So Bob, this has been great. Wonderful conversation. Delightful to talk to you. [00:40:00] Thanks so much for being on the podcast.
Bob Muglia: Well, thanks for having me, Steve. Appreciate it.