Techstrong Gang – May 16, 2024

Hey, happy Thursday to you. We've got a terrific Textron gang with a new first time guest, a new gang member here. I don't know if he's wearing his colors or not, but we're gonna be talking about big changes at the helm of AWS.

And I don't mean open source str me, uh, strange cloud bedfellows with Oracle and a, and a mystery person, as well as what? Googling, Googling, what's AI next to Google? All that and more on Textron gag.

Hi everyone. Alan Shival Textron here, uh, on Textron Gang for you. As I mentioned in the opening, we've got a, a eclectic mix of stories to cover as well as a brand new gang member.

We're gonna go right to, well, let me introduce you to Steven Dickens. Steven is one of our compadres from the Futurum group. He has, well, Steven, you could tell them about yourself.

Welcome to Textron Gang, and why don't you introduce yourself. I feel part of the gang. Allen, thank you for having me join.

So, as, as you said, Steven Dickens, I lead our hybrid cloud infrastructure and ops practice here at the Futurum Group. Been with the business, I think I'm employee number four. Daniel tells me.

So I've been around for a while, worked at a whole bunch of companies prior to that in product management and sales roles. So glad to be part of the gang. Is there, is there sort of a special sort of band Down an eagle or like Yeah, look, you get colors.

Hopefully it doesn't wind you up in trouble or anything. Well, well, you're from New York, but, um, but anyway, yes, you are an an official gang member. Stephen, welcome and thanks for being here, joining Steven and I today.

To my immediate left, Bonnie Schneider, our Echo Insights editor and, uh, sustainable it person extraordinaire. Welcome Bonnie. Thanks, Alan.

Great to be here. Thank you. And then to my far left from the Bronx, which down here in Boca for now, um, our Chief Content officer, Mike Ard.

Hey, Mike, how are you? I'm doing great. Always happy to be here.

Great. Go Rangers. Go Rangers.

What about the Knicks? The Knicks? One last.

All right, there're up three, two, but, okay. Hey, enough New York sports. Let's go over, we've got some great stories to cover today.

Mike, what do we got? All Right, well, let's get started with the changes of the leadership at AWS. Adam Zelensky says Zel Zel.

He is saying that, uh, he needs a break. He's gonna go spend some more time with the family. There's a new fellow at the helm there.

His name is Garmin. I think he used to run, uh, sales and marketing. But Steven, let's jump in here.

Um, a lot of times people read into these things and say, you know, is there something more to this announcement? Or is this just kind of a natural changing of the guard? Well, I mean, I think I spent some time looking into this yesterday.

I looked at some of the SEC filings for AWS. They didn't make a stock award to so Lipsky for this year or for last year. So if you're not making stock awards to CEOs, is that indicative?

Don't know. Am I reading too much there? I think the other piece to this is, whilst Amazon's posting good numbers, 17% in their last quarter for growth for AWS, and that business is now in au a hundred billion annually, those numbers weren't as big as Google and Microsoft's growth.

So if you are Jassy looking at this going, I need to be posting the same numbers to keep share as Microsoft and, and Google. Maybe there's is, maybe there's something there as well. So I think, I don't know, Garman, I think from everything I hear, he's, he is, he's gonna be focused on driving growth obviously comes from a sales and marketing background.

So growth and execution on top line is gonna be something he's comfortable with. It's gonna be interesting to see that play out for sure. I think.

Absolutely. Look, I got a couple of thoughts on this surprise. So first of all, in terms of the stock, look, a CEO has a special comp plan, obviously, but the, the AWS way is that when you come on, you basically get four years worth of stock.

And they're not options. They're flat out just grants of stock that, you know, you vest over the four years. So it's not unusual for an average, you know, for an AWS exec not to get new stock every year.

So I don't know how much to read into that, but if, if you believe, if you believe what was in the, uh, Andy's blog post and then Adam's blog post, right? Adam was there at the beginning of AWS in 2005, he stayed on and I think through 2011 or something like that, and then went off and, and, you know, wanted to do his own thing. And, and he was CEO at other company and has done, you know, several other ventures.

And when, when Jeff Bezos was stepping down and, and, uh, and the c Andy Jassy was moving up, they put on a search for new CEO. And rather than hiring someone from within, they reached out to Adam, who was still an insider. And, and the story says is that he was brought in to groom the next level, the next generation of AWS leadership.

And he's been there three, three plus years, whatever. I believe Adam's background is also heading up sales at AWS that when he left, uh, the, was it Garmin is the, is the new CEO Garmin kind of stepped into that role. So he was in what Adam was doing before Adam left AWS years ago.

So it could be a natural progression of, of a leadership change there. Look, it's hard for AWS to grow, you know, it's a lot harder when you do. And as, as Steven said, a hundred billion, I think they did 88 billion last year.

It's hard to, you know, grow at 25%. It's a lot easier to grow 25% when you're only doing 10 billion or 20 billion. And, and so I think that though the war of big numbers sort of catches up to 'em.

But make no mistake, AWS is still the 800 pound gorilla, right? They, they rule the roost. I think Garmin was one of the first employees at eight.

He was also there in 2005. Yes. So he's been around.

So there's a fair amount of continuity there. The one big change immediately was, I guess Adam had been personally running sustainability issues for AWS Yeah. That got transferred immediately to some other, Yeah, the same day they made the announcement, um, later, um, Amazon CEO announced that sustainability will move from something that reports directly to the CEO to community impact and global communications.

So that's a big change because Amazon's been a leader in sustainability and with AWS and their sustainability partners. And that's growing as well because you have climate tech and, and energy efficiency all under that umbrella. So Steven AI is all the range.

And of course, part of the problem with AI is that, well, it kind of consumes a lot of resources and generates a lot of carbon. So is there a, uh, a, a correlation here or is this one of these incidents where it's just kind of a happy accident? Well, I mean, I think if you, so we've got obviously a WSA hundred billion, Amazon's a big business outside of AWS and they've got a huge footprint with warehouses and distribution network.

So I think it's fair that the ESG should have its own sort of tier one position in that leadership team. Yes, it's gonna have an impact act into AWS and you're right, there's gonna be a large focus on kind of the AI sort of wave and the how hungry those machines are. But Amazon, at a group level, thinking about ESG beyond the data center, they're thinking about electric vehicles and their partnership with rivian and they're thinking about smart manuf, smart warehouses and how they do things there and supply chain, it's a bigger beast.

I think you do make a good point around AI being concrete. These GPUs and these GPU farms are gonna put huge, huge constraints on data centers. We're already seeing, you know, I saw something come across my desk the other day.

The data center providers are looking to put in nuclear power plants next to their data centers. We're gonna see some weird and wacky stuff over the next four or five years, you know, around hydro, around putting these data centers in cold parts of the world so that they can pull in natural, sort of local environmental to be able to call these things. Yes, that's gonna be a part of the equation.

And obviously if AWS is running a hundred billion dollar cloud business, they're gonna have that problem probably more than anybody else. They're all gonna be in strange places. Do you think Amazon is suffering a little bit of AI envy when they look over at open AI and Microsoft and some of these Other No, I, I, you know what?

I think they're making their bets like any smart investor does, and they're not, they're not dumb people there, you know, they have a significant, I dunno if it's a significant, but they have quite a stake in philanthropic. They, they're developing their own internal AI resources on top of that. I, if I'm not mistaken, they are the biggest consumer or buyer of the Nvidia GPUs.

Right. But I think they're also developing their own, uh, AI chips. And, and you know, Steven, to your point about data center builds, yes, I, you know, the knee jerk reaction is build, build, you know, drill, baby drill or whatever you want to call it.

But I think, I don't know if that's smart money. I think that's chasing now money. I fully think that as a result of ESG efforts and these kinds of things, we're gonna come up with a better mouse trap to do AI computing that's not as resource intensive.

And before I would go invest in a 12 year plan, 'cause you know, how long does it take like a nuclear plan, right? To go get licensing from the federal nerc, FERC folks and, and all of these things. And the time you build it, it gets opened and it's running it's seven to 10 years minimum.

In seven to 10 years, I think you're gonna have much more efficient chips that'll negate the need for that. So I, I think that is, you know, smart business people weighing short-term, long-term. How much do you wanna put into that kind of stuff Or heaven's, Well, Alan, you make, you make a great point.

You look at, I mean, AWS has been investing in its own custom silicon with in and train we're on, I don't know what generations of those, but they've been at them for a while now. You know, it's not something that they've just announced. They're on multiple generations of that.

So I think they firmly believe that training and inference need a better mousetrap to use your words, and they're already investing in that. So I think yes, they've got a huge strategic relationship with Nvidia and everybody needs to have one of those. But there's also optionality, you know, so not everything needs a H 100 in order to be, um, sort of in the AI space.

There's a lot of stuff you can do on custom silicon, such as in and train that you don't need to have an Nvidia GPU to achieve. Right? I think you're right about that in the sense that we are only using the GPU because we need parallel processing for training.

But there's gonna be, I think the software guys are gonna actually have a lot to say about this in the months ahead, where they're gonna say, you know what? There's more efficient ways to do this and, and we'll include a mix of processors. But I would just say, don't forget those developers.

I got more money on the software guys than the hardware guys. Well, Look, thank all of us are of age except you bond. Uh, we're, don't Throw me into that.

It's not fully Gray yet. Yeah, I get you. I get you.

But it's not fully gray. But, but the story of our careers has always been that software has displaced hardware eventually, right? First we do things in silicon and then we learn to do it in, in software, basically.

Will AI follow that pattern? Or is this truly the re you know, 'cause in many ways, look, thanks to Nvidia and ai, hardware is back baby. Yeah, right?

It's, it's a subtle mix, right? Some things we drop into the chip because they're common and it's better to run them there and it's more efficient from a speed perspective. But other times we write better software to make better use of the hardware.

And I think to your point, it's more often the case that software carries the day than the hardware guys. 'cause it takes them, I mean, I'm just thinking back in time when the virtual machines came out. I mean, how long was it before Intel got around to optimizing instruction sets for virtual machines?

I think I set my watch by the like seven years. Well, they, they had a reason not to too. I mean, they didn't, they wanted everyone to just keep screwing servers it, but, uh, well, I mean that, we're Gonna talk about Google coming up, but Google announced trillion yesterday, which is the sixth generation of their TPU.

I think you're right on that cycle. It's just where we are in that cycle. I think we're seeing, you know, new hardware companies come to market.

We're seeing new processor types. I'm seeing more innovation in silicon than I've ever seen. No, no doubt.

So I agree that that cycle is gonna happen. I think it's just where we are in it at the moment. We're probably in the hardware part of that cycle.

It's all about Nvidia. It's all about custom silicon. You know, we're seeing companies like Grok grow massively.

I think I agree with you that the software's gonna come and eat that for lunch, but it's just when, right. I think that's probably, maybe a couple Of years if we Steven, we, we'd be rich on some island somewhere maybe making a deal with Oracle Cloud. So Wait one bet on that.

Go ahead. It's hard. We Steven gonna make a difference before or after 2030?

Well, I mean, I think we're in the fa I mean, just going back to what I was saying, I think we're in the phase of hardware is where the game is at for ai. It's gonna be the next stage of the cycle is where software's gonna be. One thing I've learned is you can't look more than five years out in technology.

Who the heck, nos. Lucky if nobody can go two, three anyway. Well, 18 months ago, none of us were talking about chat.

GPT. Absolutely. Or chat GPT-4.

Oh that came out and Yeah, and it's, I played with it yesterday in my chat. GPT account, open AI account and uh, it's pretty cool. Now does Google have four oh MV?

We'll, we'll talk about that. But first we're gonna take a break. We're gonna be back here.

We're gonna talk about some strange cloud or the cloud makes for some strange bedfellows. Stay tuned. com is the leading resource for news analysis and education on challenges facing the cybersecurity industry.

com covers all aspects of cybersecurity, including data security, DevSecOps, cloud security, application security, network security, security threats and more. com has the largest selection of security content featuring breaking news, blog posts, podcasts and more. com to learn more.

com. Home of security bloggers network. All right, and as promised, we're back and we're talking about this, uh, expanding romance between Elon Musk and Larry Ellison, where, um, the folks at, uh, Elon's Company have been using Oracle Cloud.

And I can't help but wonder if that's just because they can't really use anybody else's 'cause they're, or with everybody else. But we'll see how that goes. But, um, it's a huge deal.

10 years billions of dollars involved Alan thoughts. So let me, yeah, let me, let me just frame the deal. And, and by the way, it's not a signed deal.

It, it's more of a rumor at this point. com. The storming goes that Elon Musk knew fledgling AI company, which I guess he thinks will do well if he wins his lawsuit against open AI for open AI being too commercial.

But Elon's not gonna be too commercial. I'm sure he is doing this for the common man, um, is making a deal with Oracle Cloud to use up to 10 billion with a b dollars worth of Oracle Cloud, GPU resources for Elon's non-commercial ai, uh, business that he's generating. Assuming Elon raises the $6 billion in venture capital and we'll see how much of his old check he writes into it raises $6 billion to get this thing off the ground so it can compete with the likes of OpenAI and Microsoft and Google and, and all of the rest.

To me, I look at this and I say, look, this is fomo. This is pure fomo. You got, you got two players who are not big players in the AI space who wanna be, and so we're putting them together.

Stephen, go ahead. So, So I'll give the flip side of that. So imagine if you had a platform with 4 million vehicles driving around the world, each of them with eight cameras on them being able to take real time data around what's happening on streets, pedestrians, road markings, what's going on in the world.

And you were able to bring that back to a Dojo supercomputer where you've built your own chips. Maybe that'd make you think about Tesla in a slightly different way than being a car manufacturer. I wrote about this in Forbes a few, few, uh, months back.

Think everybody's thinking about Tesla as a car manufacturer. They are a million miles away from being a car manufacturer. We look at Google Maps and how those ugly cars that have to drive around and get pictures of Elon's got a fleet of 4 million plus vehicles driving around every day capturing all this information.

You deep dive into the Tesla, Ts and Cs and I have all of that data from your cameras is going straight back to um, build full self-driving capabilities and also a huge corpus of data that you can do things with. So I think about Tesla of AI's gonna be about data sets. Who's got the biggest corpus of data?

You know, Google with YouTube, you've got Microsoft with and, and Google with G Suite. You've got Microsoft with what they've got with Outlook and all those big corpus of data. Who's got the biggest corpus of data of driving around the world with vision?

It's Tesla. So you pull that back and you try and think about compute v computer vision type work, use cases, how people navigate around the world, how people walk, how people cross the road, how cars drive that huge corpus of vision data. Tesla's got a lock on it.

So I don't think it's about fomo. I think they've already got that and they've been doing this Dojo suit computer for years anyway and probably got one of the biggest Nvidia GPU forms today. So this isn't, there's no there behind for a specific use case of Robotaxis and us navigating out in the wild.

Tesla's got the best corpus of data out there. And Larry and Elon, they're bo, they're bros, their friends. I got the opportunity to ask Larry some questions at a recent Oracle event.

He was effusive about false self-driving guaranteeing that Elon's gonna get it done. Which I, he didn't need to make that comment 'cause we were talking about a database launch, but he unprompted to talk about false self-driving. So it doesn't surprise me that those two were in bed together about a cloud project and just maybe I would frame it a little bit different.

That's a vision I can not see. I wouldn't, I so, so, so I would argue differently though, right? Yeah, I've got that in my brain now and I'm gonna live with that.

That, So couldn't I just easily replicate that with a Uber plus Amazon plus a couple of car manufacturers and I'm good to go because I'm gonna get the same Data. So I, I think you need Economics. You need economics, Mike.

Not only that need, that's what you commission. Yeah. You need, there's the same problem we have with the LLMs.

No, they're using data that they didn't have permission to use and, and now in the case of Tesla, uh, Steven says people did sign and they didn't notice, they signed it to Ts and Cs. That basically allows Tesla to suck all that information back up. Um, and look, I never thought of Tesla as a car company either.

I think they're a battery and software company wrapped around a car, right? I I don't think their cars necessarily are particularly well made or, or great. But their battery technology is top of the line and their software is heads and tails above anything that the big three or worldwide car manufacturers have.

Or it was, I think people are catching up, but I don't know if this is truly about Tesla or if this is about Elon Musk ego that he's not part of open AI and he wants an open AI competitor. 'cause there's more to open AI than Google Street View or Way Pro or all of it. I mean, I mean there's some weird ownership of OpenAI being a not-for-profit and then you transfer into a profits for-profit business.

Elon was an original investor in bootstrapped OpenAI when it was an not-for-profit. So I think there's some weirdness there. And that's probably Elon just from an ego point of view and from a personal perspective thinking that's just weird and that shouldn't be the case.

I think you've gotta detach that from what Tesla and his business interests are. Elon's got enough money to throw lawyers at something that he thinks is weird and shouldn't be the case. I think that's the open AI piece.

I think his business thoughts are, I got a massive corpus of data. AI's gonna be vital for full self-driving. It's gonna be vital for what is doing with SpaceX.

So I think you've gotta him as a person and his ego is probably prosecuting the open AI law lawsuit just 'cause he thinks it's odd and he thinks he was wronged. You can judge that as you want. His business side of his brain is probably thinking, I've gotta get an AI platform.

You know what Oracle's doing? They launched 23 AI as an open as a database. I got chance to sort of get, um, pre, pre pre-brief on that.

They're doing some great stuff. I think Oracle, OCI the cloud platform's a great platform for data. So, you know, and he's got a relationship, probably a really close relationship with, um, you know, with Larry that he doesn't have with the, with AWS and he doesn't have that with Microsoft.

No, I mean, so I think you've gotta detach the business from the personal, It's hard to do that with that individual. But let, let's focus on the Oracle piece of it. Look, Larry's sitting out on Lanai, you know his island out there, there's Nashville.

Yeah. Then they just, well, Oracle's moving Nashville, not Larry, Larry, Larry lives on his island. But, um, and this is a huge deal for Oracle.

Oracle. Look, they do have a great cloud, especially for data, right? You, as you would expect from the database company, um, they are well behind AWS and Google and Azure and they don't wanna be that Larry, you know, Larry playing, if you're not the top three, get out and Larry will keep chipping in there, you know, chipping at that rock till he is in the top three.

And you know, this is a $10 billion deal. That's 10 per, well, it's spread out over 10 years. So it's not all at once, but it's still a sizable deal.

It's a marque name. Um, it's, it's a great thing for Oracle clout, right? Because too, for too many people, they think of the big three and they, and they leave Oracle out, right?

And look, the fact that we're talking about Oracle and not an IBM cloud is probably music to Larry's ears anyway. But you know, it, it's a, it's a great deal for Oracle Cloud. Assuming they get these GPUs, I'm not Sure that the data they have is as rich as you might think.

'cause most of it is in a structured format and that's like a sliver of the data you need to train AI models, which is all unstructured. I don't think Oracle has a lot of unstructured data. I think most of what they got is your classic relational database stuff.

And we've seen databases run away from relational databases for the last five years where everybody's Abs on SQL databases like Oracle. Yes. Oracle has A-A-A-A-A fire in every or Market leader in any of those other categories.

Well, they're looking well, they're moved to Nashville. They focused on health and healthcare is huge. So they're targeting that with data and ai.

There's a lot obviously they can get that. So, you know, can we hook up the HE healthcare data in the car you look At? Yeah, Well you look at, you look at Cerner, number one, number two, you know, against Epic from a healthcare perspective, massive corpus of data for one of the most explosive areas that's gonna happen with ai.

Going back to what you were saying, they've, they've got a, my MySQL play. So as Unru structured, they made that acquisition that came through. So yes, we've got this traditional view of Oracle that they're the database company.

They are still, but AI's gonna revolutionize healthcare. They just bought the largest, you know, one or two, I don't know the market share, but it's certainly number one or two of, um, electronic patient record systems. And that's a huge corpus of data you put in the vision stuff with Tesla and that relationship.

And I think, Alan, the other thing I would say is they're number three by market share and number four by market share. OCI If you were to define the market slightly differently and say Enterprise and you took out Office 3, 6 5, and G Suite, are they as far behind? Probably not.

Yeah. So if You talk about Enterprise Cloud, Yeah. Rather than consumer cloud, if you will, and sort of end user apps, you take out G Suite and Office 3 6 5 and they never report those numbers independently.

You take them out of the, the cloud numbers, it's probably a different story. They, they're still probably fourth place, but they're not as far behind AWS S. But you take out 3, 6, 5 and and G Suite from the other two and it's probably a lot faster.

I I don't disagree there, I don't disagree there. But I mean, to give Microsoft and Google their due, they built their cloud offerings around us. Those were always meant to be the engines there that, that, that kind of drove it.

Right? Uh, what's interesting is, you know, look, we talk about data and, and healthcare, right? Estimates are that as, as we move fully to electronic healthcare documents, something like 40 to 60% of all the data stored attached to the internet will be healthcare related.

Whether it's MRI films or, or you know, other types of medical, uh, uh, data. It it, you know, it's like early on in the internet, 40% of every bit running through the, the pipes was poured. Now it'll be healthcare.

And that pretty much sums up where we are in, in society. So, but so, so if I have a heart attack while driving Oracle will dispatch an ambulance. There will no likely let them, they'll dispatch A Tesla ambulance come and pick you up my Automated drive ambulance.

Oh Goodness. Yeah. An automated, a full self-driving Tesla ambulance will come and pick you up.

Good stuff. Good stuff. All right, let's move on From our friends or of Larry.

We will continue this saga of Larry and Elon, perhaps at another later date here on Text Drug Gang. But let's now we're gonna take a break and we're gonna come back with the great googly Moogle. Google announces a whole raft, a whole new raft of ai, real make believe.

I will discuss it. You are watching Textron Gang, Welcome back. In our final segment we're talking about, whoa, what's going on with Google?

They had their big Google IO conference this week and they announced all kinds of stuff, none of which is actually here and now there's a prototype of, uh, AI assistant that can talk and here help you find your glasses and debug code, they say. And then there's all kinds of new models coming down the pike that may be available later this year, early next. And there's even new hardware.

What's going on with Google from your perspective here? 'cause it seems like a lot of this stuff is either I, I'm gonna be a player or, you know, as soon as I get off this couch, I'm gonna kick some butt. So Google announces more AI names and offerings.

Wait, let me check my list of how many Google AI offerings and names I've seen in the last year. They all start running in, they, I mean, whoever's making the names up, it's got a great job security over there at Google because I'm, I'm losing track. I can't keep track of all the different things and none of them seem to stick.

You know, we talk about fomo. I I don't think it should be lost on anyone out here that at the, you know, the same week that, uh, open AI announced open AI four oh as, uh, chat GT four Oh as opposed to chat GT four, which does some of the things that you just, you know, Google, is it oram or, or, or RA something? Um, GT four, oh, I think no, GPT-4 oh is the open ai.

But, but that has, you can speak to it and it talks back to you and all of that as well. Um, I, I think the thing with Google is I think they're shooting themselves in the foot by just announcing too many things, too many things that aren't even available or real yet because they want to be perceived as keeping up with the, you know, the, is meaning open ai, Microsoft. And it looks like this unholy alliance now of open ai, Microsoft, and potentially Apple.

Right? Go. That's Google's worst nightmare.

Mm-Hmm. Um, and, and so they gotta show that they're keeping pace. So they're, you know, they're announcing this stuff.

When is it gonna be real? Will it ever be real? Will it be any good?

Who the heck knows? Steven, what do you think? So, Uh, maybe take the flip side of that Google io as an event's been scheduled for a while.

All those announcements have been worked on for the last six, nine months And maybe open, open PI stole their thunder. I mean that, that's possible too. So chat GT comes along.

It wasn't by accident that they'd decided to announce it a couple of days before Google io. Google io has been. So it's, I don't see this as Google reacting to chat GPTI see it as chat GPT trying to get ahead of Google.

So all of these guys are trying to one up each other. AWS wanted to brief me the Thursday before I went out to Google Cloud next. Why did they want to brief me the Thursday before I flew out on the Monday to Google?

'cause they wanted to have AWS's stuff fresh in my mind before I went out to Google. So they're all trying to one up themselves and one up each other and get their stuff out early. The Google io sort of trained would've been well established.

They would've been planning that for six months. I was surprised by how much came out considering Google Cloud next was, what was that a month ago? Mm-Hmm.

You know, the TPU Yeah. Trillium stuff. I mean TPUV six is now gonna be called Trillium.

I love, I mean, as you say, the person naming this stuff has got a great job right now. What was wrong with TPU? But, so I've gotta relearn some new words.

Trillium, um, I don't think, I don't think I'd read into this that it's Google reacting to chat. GPTI think it's probably chat GPT reacting to Google because Google, I was a, a staple in the car in the calendar and that's been locked in. So maybe I'd see that a little differently.

But the difference is chat GT or o is actually available right now. You could go on and use it today. Yeah, I've been playing with it.

It does a bad job of describing Daniel Newman wearing a j an arsenal jersey, by the way. But I tried that one. I saw that it gave him a, it gave him a full head of hair.

Yeah. Which is obviously wrong, but I mean, I think we're on the, the, the innovation cycle is so fast right now. Yes.

Everything is kind of a three month turn. I think Google giving a state probably what is more a statement of direction than a product launch, you know, with these things are gonna come. Uh, I dunno, I don't see that as a bad thing.

It's, it's gonna be tough. How Do you think this is gonna play out? 'cause what Google's really saying is that there's gonna be kinda one master AI assistant that we speak to that will find your glasses and debug your software versus, you know, we've had this conversation on a couple of shows now where everybody's saying, you know, well here's my, here's my copilot.

So do I need all these other random copilots or will there be, you know, one or two that are gonna do everything? That's interesting. I don't know Bonnie.

Well, you look At, go ahead Steve. AWS has got Q as its assistant Uhhuh. Microsoft's got co-pilot.

Google's what got what it announced yesterday. Gemini. Gemini, yeah.

Yeah. And Trillium that you, you mentioned Steven, I think Google's put an official name on it, the LLM. But they're just showing a prototype something thing that they have a project called, it's called Astro or something like that.

That's True Astro. Yeah. And it's, it's, it is, it's interesting watching both announcements because it gets you so excited and then you keep reading and it says, oh, it's actually not available for the next few months.

Um, for some of them. But one of the things I just wanted to mention about Trillium in their announcements that Google is making and and across some of the other ones is that energy efficiency was, it mentioned in many of these new announcements that Trillium, for example, is 67% more energy efficient. And other Google AI models are boasting that they're able to be more energy efficient by only using part of the model at one time.

They're, they're, they're able to kind of designate areas of resources. I mean that's, I I don't know that's what they're saying, but That's, but that's the, but that's good. I I think that's evolution, right?

You, you learn to, uh, build boats efficiencies in shut down what you're not using. Right. More efficient.

I think that's, that, that's the way of things, if you will. And, and, and it's a good thing. But Mike, to your point, look, the Microsoft copilot was primarily, you know, first sort in GitHub.

Uh, we are gonna see it in those Office 365 or whatever they call 365. Now, um, interestingly, I downloaded Copilot for my iPad. Pretty cool.

I don't know if any of you have played with it. It, it works really nice. It's, it's a good AI thing.

But I, I think what your point is, are we gonna have an Uber, no pun intended to the car taxi business, but we can have an Uber AI who kinda wipes our nose for us and cooks dinner and does everything else. It's Mama Mm-Hmm. You know, or are we just gonna have a series of very specialized ais that reform very specialized tasks for us?

And we probably all of the above, I think we're gonna have an AI throw down. We're gonna have, the bots are gonna like battle it out and they Think movie like that. Like the guy was, uh, who was it?

Uh, Hugh Jackman was the manager of the, of the bot that, the fighting bot that was kind of, yeah. Beat up and stuff. It, that was a good movie.

I think it's gonna be in which ecosystem you're gonna be in. You know, Apple's gonna have its AI meta has already started rolling out AI across all of its platforms now. We've seen that Google's got its ecosystem with YouTube, with G Suite, with, you know, with what they do on the phones with Android, there's gonna be a number, depending on which ecosystem you spend the most time in, you're gonna probably gravitate to that.

So if you are an Apple fanboy, you are gonna be in that ecosystem. If you are an Android person, you're gonna be in that ecosystem. If you live your life in Microsoft teams and Microsoft applications at work, you're gonna use that depending on what, and you're probably gonna have two or three of those.

You know, maybe your work persona is working with Microsoft and your personal personas working with Apple, you know, or you're in the Android Google ecosystem, depending on kind of where you work place is, whether you are a G Suite or a Microsoft person. You know, I think there's probably gonna be three or four of these big platforms. I don't see 20 of those big platforms.

Yeah. 'cause it's a network effect. I I think Amazon will be in there too.

Don't count those guys out. I think, Yeah, I mean, what they're doing with Q is fascinating right now that, Well, I'm not even talking AWS I'm talking like Alexa and all of that because really let, let's be clear, your general AI kind of thing you're talking about is what we wanted Siri and Alexa and Google Home and all these things to be, we even wanted clippy to be that back in the day. Right.

Going back. But, um, and I think they are, they're gonna be much better than what we have. Steven, do you think that we will get the interoperability issues right, because today it's kind of a pain in the butt.

My house, they, all the women are on Apple. All the guys are on Google, uh, and nothing Is it really, really far in your house? Yeah.

And, and nothing is interoperable. So, you know, everything's kind of a pain in the ass to like connect to somebody on video through Apple. Like Google Don't know it is in your house.

Yeah. So Steven, is it like that in your house? I can't imagine a scenario where I'd be the green bubble person in a blue bubble world in My house.

Oh, I know that. Something's wrong over there. All all, you the king of your castle all here, all, all the boys are on Android, man.

You swear by it. Well, there's only One boy in my house and That's right. And you've gotta go.

If I had a green bubble, forbid you don't up to get my feedback down, but, um, but we gotta go like download something like WhatsApp that we can all actually use that, you know, well, well, I Think Mike, that's the point for me, that's the point for me. There's gonna be these bigger ecosystem. There's gonna be an Apple ecosystem, there's gonna be a Google ecosystem, there's gonna be an Amazon ecosystem, but there's still opportunities for people like WhatsApp to be able to be that connecting layer between those two ecosystems.

So there's, whilst I see three or four big players emerging, and this is where the startup ecosystem plays a role. There's gonna be these firms that pop up that go, you, we need a messaging layer, we need a communications layer, we need a social media layer. We need a a something layer to, whether it's a home layer or a, you know, there's gonna be like Nest was back in the day.

There's gonna be these firms that kind of come together that bring the connection points across these big ecosystems. Absolutely. I'll give you an example.

Slack. They got Salesforce behind them, right? Runs ubiquitously on everything.

Mm-Hmm. Maybe we'll see that. Anyway, we're, we're outta time for this gang, uh, episode.

We will be back not tomorrow, Friday because we don't do as matter of fact. Well soon we may be doing Friday shows, but not tomorrow. We'll be back Monday with a, a fresh text Drunk gang as well as more content.

But that's not all for Text Drunk TV today. Immediately following this, we have our full text Drunk tv, uh, schedule on you. If you're on Text Drunk tv, you'll see the playlist there.

If you're on any of the tech drunk sites, your playlist is there. You can click to it right now. You don't have to listen to me Say Goodbye.

Um, or you can go to Text Drunk TV and watch any of our shows and interviews and content on demand on our YouTube channel or however you like Consuming Text Strong. Um, until then though, Steven, what a smashing debut for a new gang member. I feel like I need a bandana and I wanna Be on the show.

Something we gotta get you. I feel like you made your bones. I need a tech strong T-shirt.

That's What I need here. If you come on down, I promise you, without waiting for you. Uh, thank you.

We hope to see you on a actually, you and, and some other, uh, analysts will be with us on Monday show, I believe. Yep. So looking forward to that.

We're gonna do sort of a Laverne and Shirley meet, mark and Mindy, if that doesn't make me too old. A mashup. A mashup.

That's Mike tells me they're called. Uh, but thank you Ronnie, thank you as always for being on. Okay, Mike Ard, welcome and thank you.

Um, and most of all, those of you out there, thank you for watching and hanging out with us today. Stay tuned for the rest of our tech trunk TV programming. Until then, I'm Alan Scheel.

We're out. Cloud native now is the Web's leading resource for the growing cloud native ecosystem. com is your destination for news, thought leadership, features and webinars on cloud native architecture, Kubernetes serverless, cloud native application development, microservices, service mesh, cloud native security, and more.

Stay on the cutting edge of modern application development at Cloud Native. Now.