Transcript: Ashmeet Sidana - Founder of Engineering Capital on Investing in Technical Insights

Host: Ashmeet, welcome to the show.

Guest: Hi, Natraj, good to see you again. So before we talk about you know, venture, your career and some of the interesting things that you've done, to just set the stage to the audience listening. Can you talk a little bit about your early career in tech and how you ended up in venture capital?

Guest: Sure. I am entirely a product of Silicon Valley, even though I grew up in India and came here to study eventually to Stanford.

But I've done many different jobs so geographically very limited but different types of jobs including working as an engineer at Silicon Graphics, starting and running a company SSI early in my career where I was founder and CEO, and then director of product management at VMware when VMware was a startup.

So small early stage companies very different types of jobs. And then accidentally I became a venture capitalist.

Guest: After VMware was sold, I was thinking of doing another startup started chatting with some friends, met the good folks at Foundation Capital and Mike Shue, Catherine Gold, Bill Elmore, these people changed my life.

You know, they just said, hey, do you want to be a VC? And I was like, that's an interesting question, why do you ask?

And so one thing led to another and very grateful to them because they brought me into the business, taught me how to do investing and eventually launched my career.

Host: What was it like, you know, to be in product management? I think you were developing ESX at that point of time, if I'm not wrong. What was that experience like?

Guest: Yeah, so I think VMware was a very well run company. I was running product management for ESX server as you noticed, and ESX server became the most important product for the company by far.

I mean it really is the heart of what made VMware the tremendous success that it is today. And um it was a classic product management job in the sense that I was the quote CEO of the product where I had the responsibility but not the authority.

And so it was management, you know, by by uh by influence management by learning, doing, helping and the company became a tremendous success and I'll take partial credit for that, but obviously like any big success it was a team effort.

Host: And so then you decided to join foundational capital. So what was that journey like and how did, you know, shifting from, you know, being working in different companies or being a founder change when you started working in Foundation?

Guest: Yeah, so you know, thinking as an operating executive in any role in a company is very different from thinking like an investor in a venture capital fund or any form of investing.

And that transition in my case fortunately was a gradual transition as I mentioned I left with the intention of starting another company and doing a startup.

And so I first was exploring ideas, thinking about doing a startup as an EIR and entrepreneur in residence, but then I switched to the investing side and learned that many times you know, things that are positive in operating can be negative in investing and vice versa.

So the simplest example I always like to tell people is that you know, if you make a reference call on someone and the first thing they say is well, this person is really hard to work with.

Well if you're looking to hire someone that may be, you know, something that gives you pause that is unlikely to give you pause if you're a venture capitalist. Often some of the best founders can be very hard to work with.

Not all of them, not necessarily, but sometimes they can be. So it's a different way of thinking, it's a different way of evaluating and looking at the world and in my case it was a gradual transition.

I had the luxury and privilege where I would go to board meetings with Catherine Goore, Mike Shue, we would sit, we would analyze the board meeting, we would analyze pitch decks from entrepreneurs and that's how I learned how to invest.

I mean truly in the apprentice model is how I learned how to invest.

Host: What are the some of the investments that you made there?

Guest: Made a whole bunch of investments.

For example Azure Power is a company that I led the seed round in served on the board of eventually went public and traded at you know, over two billion dollars. so created a lot of value in terms of an investment.

Free Wheel is another company where I let the investment and most recently you may have heard about Tubi TV which was acquired by Fox. I was also the seed investor there yeah had served on the board there.

That was my last investment at Foundation Capital. So this is a very long dated business, you know, the last investment I made actually exited only about I think about a year or two ago that Fox bought to Tubi.

Host: Yeah, yeah. I I I I noticed that you let the investment for Tubi and I was about to ask what what was sort of the insight then back then on investing in Tubi?

Guest: Well, the original idea was completely different from what it became. you know, which is not unusual in company.

So Farhad who is the founder and CEO of Tubi, he had noticed that smart TVs were becoming important and there was this hypothesis that we would be able to own some piece of that software between the smart TVs and how content gets delivered to those smart TVs.

We did not think of ourselves as an ad funded Netflix competitor, just sort of the way I describe Tubi. You know, what eventually became the successful business model. But that's what good entrepreneurs do. They evolve.

You know, they change very quickly, they rapidly understand what the market is doing and they take the whole company with them. So in my book it's not a bad idea to evolve.

Guest: Even Azure the first company that I mentioned, original idea was we were going to do retail solar. It became a completely different company by the time it went public.

So that's very common in successful companies that you go through a fairly dramatic change along the way.

VMware was a test and dev tools company, we became an enterprise server consolidation business, you know really an enterprise data center infrastructure company by the time we became a substantial large business.

So those types of evolutions are normal, natural and part of what successful stories are built with.

Host: So after Foundation Capital at some point you decided to start your own fund. What was sort of like the motivation behind that?

Guest: You know, that's a really easy question to answer and a really hard one to answer depending on which day of the week it is.

The simple answer is, you know, why does someone leave a company like Google or Facebook or Microsoft to go start a company, right?

I mean you have great jobs, you have a great position, you're obviously on something which is going to be there forever and yet great people leave these companies and go start, you know, their own organization.

So there is something deep inside us, there is something, you know, very hard to explain by me as an engineer. I think it's something for psychologists and people with other expertise to explain.

That urge was certainly there and was certainly part of the story of what made me start engineering capital.

The other part of the story is that there was a very clear market opportunity which was becoming obvious to me by that time in the year 2013, 2014 when I was started thinking about this very seriously.

Um it's almost you know, eight nine years ago now. It and that was the change that was going to happen to the structure of the venture ecosystem. So venture traditionally had been a single formula that was applied again and again by the best firms.

You raise funds in the 200 to 500 million dollar range. You are a partnership of three to 10 people and you try to do series A. That's what everybody did.

That's what Secoya Excel, Greylock Foundation, Mayfield, Kleiner, everybody was trying to do that and that formula was predicated on the belief that leading the series A's where is where the value got created and you know, there were a couple of exceptions, there were a couple of growth stage forms which would do some things, but then you really went to Wall Street, you tried to go public as fast as possible.

All of that changed dramatically, you know, about 10 years ago, about 15 years ago it started changing. And those two big changes that came in were number one, the cost to start a company went down dramatically.

Guest: Number two, the cost to take a company public went dramatically. And these two opposing forces ripped apart the venture ecosystem. And that's why you saw this Cambrian explosion where firms like Secoya decided to become big.

Firms like Benchmark continued to do what they do, what they were doing then and firms like me got created who decided to focus on the emerging seed ecosystem. So this structural change was you know, it's a one time change.

It happens in industries when technologies etc drive these forces and and that's what I took advantage of to start engineering cap.

Host: Why why did the cost of taking a company public go up? Or what do you mean by you know, cost of company? Is is it because you have to pay more you know much larger to the investment bankers or what essentially changed?

Guest: Yeah, that's a really good question and I didn't word it exactly right. It's not the cost to take it publicly. I mean that did go up a little bit with regulation and stuff and yeah, investment bankers make a little bit more.

What I really should have said is that the the typical investments that companies took before they went public went up. And the reason for that is that uh number one, it has become less attractive to be a public company CEO.

Because of over regulation because of litigation, because of the way our public markets have developed. It is not attractive to be a public company CEO.

And so people want to defer that, they want to delay that as much as possible which means they can take more capital.

Number two, there was more private capital available which could then take advantage of those disproportionate gains that come when companies go public.

Guest: People forget Microsoft went public at an 800 million valuation. Amazon went public at a 400 million valuation. These would be series B rounds in today's market.

But what that also means is that all the gains that happened from 800 million to two trillion in the case of Microsoft or Amazon or trillion let's call it a trillion dollars in round numbers um all of those gains were made by public market investors in the case of Microsoft, Amazon, Google, etc.

Um today a lot of those gains are being made by private market investors. That's not a good trend. As a citizen as someone who thinks about you know, the future of the country, I don't think that's a good policy to do.

But at the end of today I'm not a policy maker, I'm not a politician, I'm not a legislator. That's not what I do. That is what for various reasons has become the way our public markets work.

And so um it therefore, as a side effect of that, it became more attractive to stay private.

And therefore people took more money on the private markets, private markets were available to give that capital and so the total investment that happened on a cap table became very, very large.

And that's what allowed large funds to exist and large companies large firms to be built around that.

Host: So in a way if we can also say that large funds saw an opportunity that from the lesson of Amazon and Microsoft that if we can capture, if we put the company private for longer, we could capture more value is is that a absolutely that is explicitly what Secoya said when they restructured the firm a short while back when they talked about why they were going to fund as one of you know, one of their restructurings was that they wanted to continue to hold companies while they were public after they were public.

Was part of that motivation was to keep them private for longer, let them go public and then continue to hold them because there was so much value creation that occurred over there. Look, that's an individual firm strategy.

I'm talking about the structure of the industry.

Host: Yeah.

So let's let's talk about that too because I think in last couple of years this strategy sort of the whole phenomena sort of peaked at, you know, venture firms creating ever holding funds where they can hold public companies that have gone public from their portfolio like A16Z did it, Secoya did it.

I think there is a sort of thought that they're scaling back on it but how do you see that and how do you see that argument? Because you talk to LPs while you're raising your funds, right? How do you pitch that as value to LPs?

Because LPs, you know, traditionally obviously diversify between public funds and private funds, right?

So how are you sort of now you're essentially saying that I have the capacity to maintain and rebalance public market for portfolios which is not your technical skill that you've developed over the decades, right? So what is the pitch there like?

Guest: Yeah, for me personally, that's not my expertise, that's not my area that I pitched to LPs and that's not what engineering capital is about.

But clearly there are some forms you mention in recent horowits etc who have decided that they want to either build that expertise or they already have that expertise. That's up to individual firm strategy.

The person who's running the firm makes a decision about firm strategy. In other words, what is your expertise, where do you excel, where do you have a competitive advantage and how are you going to make money.

And so I think LPs evaluate that on a case by case basis. You know, certainly forms like in recent are very large and so they're talking to very large sophisticated LPs and they have the ability to evaluate that.

Host: Yeah. So you you know obviously saw this change in the industry, decided to start your own fund. Uh how did you end up on what thesis to pick? Uh not on how, but what thesis to pick in general and talk to me a little bit about that process.

Guest: Yeah, so again I want to acknowledge Catherine Gold who was a founder of Foundation Capital, you know, she was instrumental in helping me think through once I went to her and I said I really am going to start my own firm and figure out what is it that I wanted to do.

And what I wanted to find was at the intersection of where there was a market opportunity, where I had an unfair advantage, I had a competitive advantage and was something that I enjoyed doing, something I wanted to do for the rest of my life.

So where was that subset where all of those things were true and that happened to be very early stage, deep tech, technically challenging investing here in Silicon Valley and that's what engineering capital is about.

You know, it's really taking advantage of that that small subset.

Host: You you talk about technic investing into technical insights as a theme, right? Talk to me a little bit about you know what do you mean by investing in technical insights and like give me an example of what is a technical insight versus which is not a technical insight.

Guest: Sure.

So technical insights you could call them technical innovation, some people think of them as patents, some people think of them as creative new, you know, disruptive ideas, I define them in the following way because I only do software.

So I also want to distinguish myself from other people who are investing in hardware trying to do that.

Technical insights in software means if I was to describe what I'm trying to do, it would not be obvious to someone else a good engineer how you would do it. They'll go, wow, that's a difficult problem. I don't know how to do that.

So that's a technical insight. if you have a solution like that. What are some examples of technical insights? let me start with some hardware examples because those are the easiest to understand.

Today it is clear that an enormous amount of physical innovation is limited by battery capacity.

Imagine if you could make a battery, right? if it had higher capacity, if you could charge it faster, if it could last for more cycles, if it had less weight, if it took less space, anyone of those variables if you could make it better, that would be a technical insight and it would allow you to create, you know, fundamentally new business over there.

Guest: In software, it gets a little bit more subtle, a little bit more nuanced, but here are some examples. I have a company called V function.

They can take monolithic software, traditional just say Java applications, CC applications written as a single piece of software and automatically refactor them into independent microservices which can be deployed in a cloud architecture.

That's an incredibly hard problem to do. If you have a computer science background you're like well how do you deal with the global variables, you know, programs have global variables, they have state. You know, how do you split it up?

Uh that's the magic. So that's one example.

Host: Having having done splitting monoliths into microservices, it will take you years doing that. Yes. And you know tens of teams inside Microsoft for example, we have tens of people like in the cloud era example, right? We went from non cloud era to cloud era there were thousands of people just working on doing this. right it's a very big problem.

Guest: It's a very hard and a very technically challenging problem and now we have a startup where we can in the course of an hour long demo or a day long workshop, we can take your existing monolithic and show it show it to you running with separate microservices.

So, you know, that's the power of a technical insight. That's the power of a technical team saying, can we solve this problem in a creative new way that nobody has thought about before which would then unlock enormous value.

I think it's obvious to you the amount of value that a V function is able to unlock. I was going to give some other examples. I have a company called Robust Intelligence, right? They specialize in protecting machine learning models.

Machine learning when you run any kind of AI machine learning etc they are susceptible to all many forms of failure, many forms of security issues not just cyber security. In other words, in addition to the traditional cyber security.

I mean you don't want someone breaking your password or hacking your network or you know, getting access to the machine, just the model itself, the way machine learning works, you know, it's susceptible to data poisoning, it's susceptible to hallucination, it's susceptible to there are many examples of things that it's susceptible to.

Here we have a team your own singer, you know is a PhD from Berkeley, he was a professor at Harvard. He left to start the company. We have some other world's best technology uh to protect machine learning at scale.

So those are examples of technical insights in software where somebody is able to solve a hard and interesting problem.

Host: So when you first you know, started the fund one of the typical problems you faces how do you get in front of you know these good founders who are technical, uh you know, if they are searching for a VC or an early stage investor and you know they'll probably search in Google or you know they'll go to the usual players.

You're you're a new firm trying to establish yourself, trying to probably source deals like how how did you think about that in the initial dates?

Guest: Yeah, so that is the heart of the venture capital business like any other startup. I'm also a startup or at least I was a startup several years ago. And you know, people have to know about who you are.

You have to go and find your quote unquote customers, in this case they're not really customers in a traditional sense because I want to give them money, but I have to go find these people and you do them in the same way that every company, you know, finds their own way.

So for example, in my case, I'm hoping that maybe people will listen to this podcast and they will say, aha, here is a VC who is perfect for him. Here is someone who that they want to talk to where he would not be well served by a traditional form.

I also do the traditional things in terms of networking and talking to people and going to events and you know doing all of those, but you have to build your practice. Venture capital at the end of the day is a services business.

It is a business where we win if we can serve our CEOs better than anyone else. They have to be convinced that you will be the best partner for them. The best CEOs, the best companies have choices. They have multiple people willing to invest in them.

And so they have to choose you as an alternative to others and I'm very proud to say I mean I obviously have built a portfolio now that I've been successful in doing that.

Host: One of the things I think you know in in the interaction that we had last time you spoke about is the physics of venture capital.

And how sort of I think we, I mean as a venture capital as an industry sort of derailed from seeing the physics of venture capital and start investing in ideas that not necessarily are venture fit in the venture model, right?

Where you can't expect a 10X or 100X or 1000X return. Uh I I I would give a classic example of like a VC firm investing in a in an other incubator sort of like living companies.

There are a bunch of companies that came up in pandemic where you know, just they would invite founders and grab something and you know, use the space to create something. and a VC firm basically invested in that company. or let the seed round or something like that which didn't make any sense to me because what leverage or you know, especially tech makes a lot of venture sense because technology itself is a big leverage, right?

You put X amount of work, you can get 10X output from the code, you can repeatedly use it. That's where the leverage is coming from.

Uh so talk to me a little bit about you what is the physics of venture capital and what happened in you know last three, four years have we seen.

Guest: Venture capital is fundamentally a long tail business. It is a business that wins and succeeds in the rare few cases where you get exceptionally high returns. That is how venture capital has always been built.

Most companies are not well suited to venture capital returns even though they may be other good they may be reasonably good investments. on the in their own right. So it's not a knock to say that it's not a good match for a venture capital fund.

In fact it's the reverse. It's only the rare few that is a match for a venture capital fund where venture capital firm should be investing.

In the example that you gave I don't know obviously the work firm that you're talking about, but what I have seen sometimes is that firms will invest in an incubator or an accelerator in the hope that it will result in deal flow for them.

Guest: So they'll put a little bit of money in just so that they can see all the companies that go through so that they could pick one to solve the previous problem that you were describing which is how do I get new companies, you know, to how do I become aware of these new companies and how do I know that I would be able to that I get the opportunity to go and see them.

So maybe that's what they were trying to do in which case it really isn't an investment. It's really an expense. You're not looking more of a marketing expense. It's a marketing cost. It's a marketing expense. So that's a different aspect of it.

From the physics of venture capital itself has not changed fundamentally small amount of capital returning in a large outcome is what drives venture capital and that's why arguably one of the world's best firms benchmark continues to do what they've always done.

It's always been a 400 million ish sized fund with you know five ish partners making early stage investments. And they've done a fabulous job they've obviously built great returns doing it. So that continues to exist.

What happened what changed in the last few years is that the amount of wealth creation became so large that people layered on other strategies.

For example we talked about layering on a growth strategy and doing later stage investments for lower risk but lower returns and so you know that makes sense if the company's doing really well you can do that and they still call it venture cap.

Now at some point, you know, you should stop calling it venture cap. for better or for worse.

But you know, there are marketing reasons and legal reasons why people call it venture capital and so that's okay, you know, that's just noise, that's just ambiguity that exists in the market.

But true venture capital in my opinion is that very early stage high risk investment where you are looking for the 10X, 50X even 100X or as you said, you know, sometimes even a thousand X can have it. That is venture capital investment.

Host: So going back to the technical insights investing in technical insights.

Have there been things that were superb you know technical insights but didn't make a good business or like didn't you know scale to a good business like initially you thought okay this is a great technical insight and this could be something huge and didn't work out.

I know like venture is full of like you know initial excitement then you know after one year, two years, you know, that's the mathematics of it like only five companies out of 100 will actually make it.

Uh but any examples of like technical insights that you thought really really worked like more than the 80% but didn't actually work out.

Guest: Absolutely happens all the time. that's why I have gray hair. Uh this is the nature of this work is that there is a high failure rate and you have to be able to sustain that.

Um I made an investment early in my career in a company called Panalogic. there was a beautiful fundamental technical insight which was building a stateless, you know, display protocol and we built the display protocol and we built a device, we got design awards.

We built a pretty reasonable business. In fact, um I as I mentioned I was the first investor and then we subsequently raised money.

I believe from Mayfield and Goldman Sachs and Con ventures and I mean there's a whole bunch of capital that was invested in the company uh and unfortunately we were not able to build a business. Yeah. So what happened? We lost everything.

That does happen. Yeah. you mentioned protocol and in in the field of blockchain you hear the word protocol being used a lot.

I'm curious what your thoughts on the field of blockchain itself and why have if I'm right, I haven't seen you make any investments in blockchain even though it's um there're definitely some technical insights for sure.

So tell me your thought process around what's happening with blockchain as a technology and what do you see uh happening there.

Guest: So I completely agree with you. There are several very intriguing technical insights in the blockchain or web 3 world generally speaking starting with the very first paper itself Satoshi Nakamoto's paper on the Byzantine generals problem.

You know, that was a technical insight. When I went to school we studied the Byzantine generals problem as an unsolved problem in computer science, right? And here was a paper which solves the problem. Now it's not a very elegant solution.

You know, proof of work is a very um inefficient solution but whatever it was a solution. pairing that with currency I think was part of the problem um where I really don't see currency as a very good use case for the Byzantine generals problem.

Not because it can't be solved that way, but because there is a lot of political and regulatory reasons why we don't want to hand currency off to an anonymous uncontrolled medium. You know, that's just not the way currencies work.

For thousands of years uh the the governments have kept a monopoly on currency. It has been illegal in various ways to mint currency. we went off the gold standard for very good reasons.

Yes, there were some political reasons also, but for good reasons we went off the gold standard and the entire theory of economics has developed around how to manage currencies for it. That's not how the blockchain works.

So um I think the fact that they chose this use case early on was odd but okay maybe there were some valid use cases for that you know and it could have built.

And then of course we got a whole slew of scammers and you know all all the bad elements that came in which come in in every new technology early on and that has just that is kind of the current state of the world.

So we don't have a use case other than currency for this interesting set of problems. I can actually imagine use cases which I think are valid and viable use cases.

You know, the most obvious one is um all contracts, you know, should reside on a blockchain. Why not? Right? I mean it's a very obvious use case.

Um and and and I think some value could be created if we did it but right now the noise was so bad, there was so much money and you know, fake wealth creation that occurred uh through the currency and the manipulation of currency in various forms that uh I think it kept the whole industry down.

That will go away, it is clearly clearing up now and so I'm bullish on the blockchain. I believe long term some use cases will emerge and this technology will live on because those insights are fundamental and useful.

Host: Yeah, I think it's sort of over by choosing currency as the first use case they over indexed on all negative incentives possible and which attracted more scamming, which attracted you know, all Pazi economics uh and I I could have all and always seen blockchain as a good solution for identity, contracts, theoretically at least it seemed like it but it never actually came into the picture because there's so much distraction with all these things.

And I think there's also like sort of you know, once you have a hammer, you sort of use it for everything uh problem with crypto where you take blockchain and put in everything.

So there's blockchain cloud, there's blockchain, you know, you name a product, you combine it with blockchain even though I don't think it's a foundational change, right? It's not like AI where you can put it in every application.

It is a probably a vertical sector in itself which will be like really useful maybe in some category solutions. And that's how I feel about blockchain.

So it's yeah as you said there's some smoke clearing, but I feel like it's so corrupted with currency and quick profit.

It's really takes someone you know genius to come out there and create these applications who's like super focused, super stoked about the problem that they are solving and I haven't seen the proof of it and the biggest negative I feel for blockchain right now is why not the big seven companies invested significantly into the blockchain.

I think that's the big negative I see. Like why why why hasn't Microsoft done something? They had some blockchain projects but they got cancelled.

AWS did a little bit of you know dabbled in it but never actually invested huge sums into it. that's the big negative because any upcoming new technology all the big seven usually if there is a opportunity to be had uh they usually are the first ones to jump in and sort of try to use it.

Host: Especially I think in last five six years we've seen that the big seven companies always try to adopt the new technology. Even Facebook tried something with I think they tried with Libra and they they sort of realized how contentious it's going to be with regulatory bodies. Uh yeah that that that's my take on it.

Guest: The takeaway for me and I agree with you is that AI is going to be in everything. The way I like to say it is that talking about software without AI will be like talking about software without electricity.

There's no software without electricity. Yeah. Um all software runs using electricity and so that's how it's going to be. It is going to pervade every element of that.

The blockchain I think is a very specialized technical insight with some specialized use cases which will emerge.

But you know, because of the way it developed because of the way the economy and the and the industry developed we're kind of stuck in in a difficult area right now. This may actually be the best time to start a blockchain company.

If you truly want to solve a real focused narrow use case, I think this is probably a great time to start a company like that.

Host: So you mentioned AI, right?

So talk to me a little bit about you know what we're seeing in AI obviously there's a big breakout you know last year because of chat GPT it sort of captured everyone's attention, everyone is you know now paying attention to A although like things have been happening in machine learning and AI for a while like those were like if you talk to any master student like the go to you know subject to choose is machine learning it has been there like that for at least seven eight years.

Um but talk to me about what's happening in the sector and how do you see the whole sector being play out and especially what you are thinking about in terms of investment perspective.

Guest: I'm fundamentally a believer in AI. Um we have been working on AI for the last 40 or 50 years incrementally improving it. And what has happened most recently is that we've clearly hit a tipping point.

And that tipping point is now with LLM's chat GPT, some of these techniques where we are going to get an enormous amount of utility and value out of AI being applied to a wide variety of problems. So I'm a huge believer in that.

However, I also believe that AI is fundamentally a sustaining innovation. It is not a disruptive innovation. So in other words, AI actually helps the incumbents more than it helps the newcomers.

So startups don't have a kind of free ride over here, it's not going to be that easy to go in and just go build a business and.

Host: Can you little bit define you know what is innovative versus disruptive versus sustainable innovation?

Guest: Yeah, so a sustaining innovation is an innovation that helps an existing company which has an existing business run their own business better faster cheaper. That's a sustaining innovation.

So let's take a simple example, you are running let's say a trucking company in the 1980s and then suddenly there's an IBM PC and IBM PC is a classic example of a sustaining innovation.

You can buy a PC and you can now run your trucking company better. You can use it for scheduling, you can use it for communication and that's exactly what happened.

Every single company in the world eventually adopted the PC as a sustaining innovation.

A disruptive innovation is something which changes the dynamics of the distribution of the industry in a way that the existing incumbents don't have an advantage by adopting that technology. What's an example of that? The iPhone. The internet.

They were highly disruptive.

So Amazon was equal to Walmart which was a massive retailer with hundreds of thousands of locations and hundreds and thousands of employees and suddenly Amazon is able to sell head to head with the Walmart when the internet comes up. Yeah.

That was a disruptive innovation. It is Walmart has still not been able to respond to Amazon's innovation because they were able to take the internet and and leverage that so beautifully. So that's the example of a disruptive innovation.

Um the question is what is AI? I believe AI is a sustaining innovation believe it or not. That doesn't mean there won't be any disruption. No, that's not what I'm saying.

What I'm saying is that every company which is smart and intelligent will adopt AI and will benefit from that which will only increase their competitive advantage. Which means if you're a startup you have to work that much harder to win.

Now because it is because it is so innovative, it will create some disruption on the edges and yeah, startup companies will be created especially at the infrastructure layer and that's why you are seeing so much investment go suddenly into the infrastructure layer of AI.

And you know, people

---