Techstrong Gang – September 5, 2024

Hey everyone. It's Thursday. In case you didn't know, you know, we've got data centers opening everywhere and nowhere it seems the witch hunt against Big Tech continues.

And what is the future of Meanings in the age of ai? All that, and more on Textron Gang. Hi everyone, it's Alan Shimel for Textron Gang.

Welcome. Uh, for those of you who don't know what Texturing Gang Text and Gang is, this is our daily talk show at the top of our Textron TV schedule. What we're gonna spend about 45 minutes talking about what we think is big news or interesting stories for the day.

And we're joined by a roving, ever changing group of our Textron gang members. It, it's not ever changing. It's actually the same people, but we rotate them in and out, and everyone gets on maybe once a week or so.

Um, we've got a great gang line up today for you. Let me introduce you to them. First of all, uh, our, our men in Silicon Valley, uh, roving editor extraordinaire, Jon Swartz, is joining us from Silicon Valley.

Hey, John, how are you? I'm good. A it's good to be with all of you.

Good, good to have you. And then we've got the, the, our Colorado Twins, uh, from from Boulder area. Uh, she is a analyst extraordinaire, uh, with Futurum Group, Camberley Bates.

Hey Kimberly. How are you? I'm doing great, thank you.

Okay. You know, you find out interesting things about people. We found out that Kimberly's husband is a big Yankee fan from the Bronx.

So extra points to Kimberly and her husband and the whole family here. I, I knew we made the right decision riding her onto the, We're gonna swap them out for one of these days. Yeah, No, it was a good cut Industry.

So you guys can talk Yankee baseball Only when the Yankees win. Take, Take, take, take somebody brave to marry somebody from the Bronx and the Yankees fan. That's all he Gotta say.

I I give her extra point. Yep. You've got some metal.

And then, um, they're also joining us from Colorado. It's our CTO and Fu of CTA. Mitch Ashley.

Hey Mitchell. Welcome. Hey, well, thanks very much.

I don't get to do this very often, but you will notice we do the west of the Mississippi has you outnumbered this time. Yay. Yes, you do.

Go West coast. Go west. Young men and women.

Um, that's okay. And you know, we are here in our book of Raton headquarters, uh, studio for Techstrong, and I'm joined by our chief content Officer, also hailing from the Bronx, Eddie Yankees fan. Uh, it's Mike Vizard.

Hey Mike. Welcome. That's how You get on this show, Right?

I will. It is a fast track. I've heard apparently, you know, there's, there's like an underground railroad from the Rocks and Yankee Stadium right down here to Boca Raton.

Um, anyway, so guys, we, we've got some interesting stories today. You know, we're gonna kick it off with, uh, data, data, data. You know, there's a good data or data, tomato, tomato.

I usually say data, but every once in a while I throw in a data, I dunno, go figure. It's data center. And, and it's data content to me to a different Side, side of the Adida.

Aidas, Do they really say Aidas? I does. Mm-Hmm.

I always see it. Okay. Anyway, we're talking data centers everywhere.

We had this article, well, actually two articles in Techstrong, ITSM. They're in the notes. Mike, lead us off on this.

All Right, well let's get started here. We have a couple of stories on Techstrong ITSM that's talking about, well, there are not enough vacancies and data centers, and people are trying to find interesting ways to build these new data centers and maybe share them. And Kimberly, let's jump into you.

Is this all because of Gen AI and CPUs and GPUs are in demand because of those applications? Or is there something else at work here and does all this portend maybe some price hikes on cloud services? 'cause supply and demand are outta whack?

Well, I took a look at the data that that was there. I didn't get, you know, I don't have the full data set that these guys had, um, to looking at, at the data center. Um, but I suspected it's, I I suspect it has to do with what's going on with the GPUs and bringing those GPUs in because they are consuming a huge amount of capacity.

Um, I'm, I'm also wondering, you know, a week ago we talked about the possibility of repatriation of some repatriation as well. Um, and so there's possibility that's going to be cranking up, but I would think that yeah, that's, that's, that's kind of what is, uh, pressing the pedal, um, right now in terms of the building out, um, of the data centers is the other piece of it, I would think is the, um, geographic locations and, um, having the, the sovereignty of data centers needing to be in, in country. Um, that's putting a, a, a crunch on the requirements, um, for build out of data centers as well.

But that's not totally what's going on in Virginia, which is kind of running out of power right now. Well, Sure. com days.

I, I helped start a, uh, what, what we called back then an a SP, an application service provider named Inter reliant. And, and part of it we, what we started as a hosting provider and then moved into a SP and we operated our own data centers. One in Tyson's Corner, Virginia, an old a OL facility that was pretty big.

Had a big one in Atlanta. We won in, in Paris, London. And I don't think we ever opened the Silicon one, the Silicon Valley one.

But the interesting thing is, back then, and this is 19 98, 99, you know, we had all these forecasts for the next 25 years of data center growth. And, you know, it was through the roof. We weren't gonna be able to build data centers fast enough.

com bubble hurt burst, and all of a sudden we had so much extra capacity and data centers and dark fiber. You know, I then started Mitchell and I, two of the three co-founders of a company called Still Secure out in Boulder, Lewisville, to be exact. And I remember driving through Interlochen in those, I dunno if you've ever been to Interlochen by the Omni resort near the behind, behind the mall, you know, level three was there, and it used to be all the storage places there.

And there was so much angst over all of these empty data centers because we were, we were never going to fill up all that data center space, it seemed, and, and all the dark fiber that we, you know, when were we gonna light that up? And here we are now, we finally, 25 years later, we've, we've filled it up. I think there's two things here, though.

Number one is the hyperscalers, the cloud, right? Your AWS, Google, Microsoft, Oracle, how many data centers have they lit up and their data centers are Mm-Hmm. Truly hyperscale.

But then I think you have the rack spaces and your, your colo centers. There's a thriving business. And you and I lit a ton of them down here in SA who are having a thriving business in co-location data centers.

And, um, and they're not small rinky dink data centers. They're, you know, they're, they're pretty big and they're everywhere. And they are everywhere.

So I, you know, as the, the, my biggest issue though is where are we getting the electricity to cool them down, power them up, turn them on, light 'em up, right? And, and you know, we don't have Bonnie here, but, uh, Bonnie Schneider, but the sustainability of these data centers, I think becomes a real issue. I think Alan, part of this is recognition around data.

We talked a little bit about this before of it's expensive and difficult to move data around. And if your core data from your whatever manufacturing or inventory system maybe running on a mainframe or some local application in your private data center, maybe it's, maybe you're thinking about doing other AppSec. There's, or maybe AI AppSec of course, too, in combination with that.

So I, I remember the time you were talking about the inter reliant and leading into to still secure and there was such an overbuild. We, we also, we saw this arbitrage opportunity of data unused excess data center space. And, uh, maybe that was the early forming of the cloud in some way, but that, that evaporated pretty quick.

I think these, these situations change, uh, significantly. And, and if cash becomes even a greater crunch, then I think we'll see less investment in local data centers if cash flow is an issue for companies. So it, it's not all a pendulum, we're not swinging the pendulum all the way back to local data centers.

I think there's some, some measured decisions about what do we wanna do locally in our own world. Kimberly, let me ask you this though. Geography plays a role in this because, uh, for a while there we were like, oh, we're just gonna build all these data centers in remote locations and they'll be next to rivers and they'll be cheap power.

But then, you know, if you look at what's going on in Virginia, we want the data center as close as possible to an internet exchange. And that's about latency at the end of the day. Um, how much does that drive decisions about where these data centers are gonna be built?

Because, um, you could theoretically put 'em almost anywhere. But the issue is what kind of data center goes where? I think significantly, you're looking at the latency.

I mean, there's a, there's a story that I think it was probably 10 years ago when, um, I believe it was the trading that they established fiber, fiber optic line in Chicago and New York. And it was a straight line. They were not gonna do this.

They were gonna do an absolute straight line to take every bit of latency out of that network as possible for the trading purposes. Um, I know of a, a mining company that had moved a lot to the cloud and then hit the wall because of the, in, because of the internet and latency problems and reliability, had to move everything out of the cloud back into the mining facilities to where they're located, where the mining is the same thing as Alan was talking about, the manufacturing, manufacturing portion of that. Um, the, uh, decisions are being made, you know, from the ITSMs in know per your article here on what is the latency that they can withstand and therefore where would this reside?

And, but on the other hand, I have another situation where I've got median entertainment. So worldwide technology, I could be filming in New Zealand. I could have developers or artists, you know, located all over the world.

And in that situation, I'm putting on global file system that's completely global, that's completely in the cloud because I wanna be able to share, you know, and make sure that I have, you know, the responsiveness that I need on my, my computer in which I'm doing the development work. So it's really application by application about where this needs to reside. In particular, as, you know, you, as you guys were talking about is, as Mitch was talking about, where's the data and where's the compute and how are these things, these two things gonna interact and those decisions are, even though we're being more and more aware of that.

And I think the AI situation has even raised the, the level of awareness. Now, Can I, can I Throw out like some, I don't know, somewhat contrarian viewpoint or just an idea or concept is how, Let's hear from the Giants fan. Oh yeah.

A few of us are left as we roll our way to the basement. Um, hey, so I, I wanna ask you about just this kind of, there's a kind of a general anxiety out here in the valley that I'm starting to see or pick up even some referring to the softening in the AI markets. And I think of a lot of that has to do with the forecast for Nvidia, which I know there are these lofty expectations, but still the, um, earnings forecast didn't meet the criteria as high as the bar is.

And then we have this law or a bill that's gonna be signed into law, or is expected to be signed into law any day now by Gavin Newsom that overlooks, um, safety issues involving large AI models. And then we're gonna talk about this DOJ investigation of Nvidia. So what I'm, basically, what I'm trying to ask is, I'm gonna throw this out to all of you, is, do you think these events and in growing concern about AI and it its ability to grow quickly, as quickly as we thought, does that impact the data center expansion?

Because I, I, I wanna allude to what Alan had said earlier about the dot Kong bust. And I, I'm not, I don't think it's anywhere close to that in terms of magnitude, but I do feel as if there's some sort of retrenchment going on out here. And Ron, Go, Kimberly, In my view, this was expected and if we didn't plan for this, that's ridiculous.

And I think what, what we've, I've been expecting to see is coming up against the wall on the problem of data and being able to report out where did I get my information, how did that information, so I regulate, they've been planning for this, or most of the companies have been planning to say, I need to be able to tell the regulators how this decision was made that was been part of their legal oversight or their advisory oversight they've had. The second piece of it is they have been learning about what it means to manage data. And that's difficult.

And we've always run into this problem. We've done data warehousing, we've done anything around data, it gets ugly. And so now we're hitting that.

And so it slows down until we get our hands around that. Once we get our hands around it, it will pick back up and go. But to me, this was an expected slow down.

We're gonna see a slow down in implementation, um, of it for, you know, and I figured we'd be hit this now and then we're gonna be kind of slowing slowish until the summertime. So it was inevitable, in other words. Right.

These other insinuating sta circumstances are just, you know, something on the side as well. Right. The investors can't figure that one out.

Well, yeah, they're, they're not the most rational people, I'll be honest. Yeah, I, I did some just quick thoughts on, on John on, on what you're saying is number one, you know, this idea of only if you had a rational exuberance to quote a phrase to begin with, right? Mm-Hmm.

I mean, like the whole thing of Nvidia losing, I think it was 9% or something, they dropped, you know, on, on a quarter where they exceeded the street estimates, they blew out the earnings, right? The, the, the profit over and above earnings as well as top line. And they still took up spanking, well, either the streets estimates were widely, widely undervalued or the irrational exuberance.

Why I, you know, what were these people looking for that would allow that stock to stay up? Or was, is this just sort of a natural sort of, uh, you know, like a correction, A correction we add to some sort of, right, right. That's, you know, I'll be blunt about it.

Trying to put like, rational thought on an irrational market is like a pointless exercise. It's impossible. Yeah.

It's, it's like, it's a mood kind of betting swing. It has nothing to do with the fundamentals of what happens with you. So I would like to come back to the conversation at hand.

Are we gonna see a situation where, um, we are gonna be over-provisioned on GPUs because we're investing too heavily because there might be a sh not as much demand or has the fundamentals of the investments are currently aligned gonna play out as we expect because the next wave of applications will come and will develop it. It's just the timeline is a little bit off compared to what Wall Street thinks. Kimberly, Maybe contrary to some of the data that Ur Gers put some data out that says this is how much this market is gonna be growing.

I think we're gonna see balance in there and with CPU being used for tr for applications along with the GPUs. So we're also gonna see some GPUs coming out from a, from some of their competitors. Um, that's gonna all balance the market out.

Um, not every application, or not every use case is needs GPUs. And we saw that with some of the testing that we've done in the Signal six five lab, um, that we was coordinated with with Dell doing some work with their servers. And so there's some applications, and I can't remember which one it was, um, that we were using, you know, more of the G CPUs versus the GPUs application areas that were out there.

So I think we're gonna see a balance, um, of what's going to be needed there. Mitch, follow up question. If you're the CTO and you're looking at what is probably gonna be a price hike on cloud services because demand exceeds supply, are you gonna move some applications?

I mean, what kind of decisions are you gonna make? A lot of your applications, maybe you put in the cloud in the first place three or four years ago? 'cause that was the only game in town.

So do you revisit all that or, you know, what, what's your wallet telling you? I I think there's a natural, you have to revisit it on a regular basis, right? 'cause prices are changing and your, you know, the economics of it's cha of it changes.

The one thing that's incongruent with, with the AI chip argument to me is if there's an arms race to buy AI chips, you know, to build out your data centers in the cloud providers, how are you, unless you're one of the ma one of the few massive enterprise organizations, nobody else is gonna be able to get those chips. Either. You, you putting them in your own data center, you have an even greater struggle supply.

And what tends to happen is, is in these lack of supplies, people over order, right? Like I, I'm gonna put a bigger order in so I don't miss that second shipment of chips. Like I did the first one.

And you, you build this backlog of over orders, over requests. And then so at some point there's a writing effect to Kimberly's point about balance of can be competitive, can be whatever, manufacturing, more capacities available. And all of a sudden people start canceling those second and third orders 'cause they don't actually need them.

That was a safety valve in case the first or second one missed it. So this is one of the, this is a bubble where we're gonna see, um, bursting and popping and not like, you know, a market crashing bubble, I don't think. But this is the over inflated expectations.

Every problem could be solved by ai. That that's where we were six months ago. And everybody thought that was, you know, we won't need developers, we won't need marketing, we won't need analysts, we won't need whatever.

We just have AI do it all for us. And now we know that of course, that none of us believed it, but we know that's not true. And that's the realization we're coming to is like, you know what?

This stuff's hard. This is not easy to just, yeah, I can use chat. GBT doesn't mean I can displace 500 people in my company tomorrow.

Yeah. Can't really one follow up to that. Um, theoretically, you know, the benefit of being in the cloud was I could dynamically consume things, but when I hear from folks when it comes to GPUs, they're signing up for like almost year long leases of GPUs and they're basically reserving the capacity for as much as they can get.

So is that kind of like, to me that feels much more like, you know, hosting? Well, well, I can't believe answer That. That's, I I believe that the Yeah, I understand that that is what they're doing.

Um, the other piece of that is we've been talking about how do you move the data towards where the GPU is, because that's difficult to do. Um, so that, that's another, um, part of the pipeline kind of piece of it. But yes, that they are doing that, um, particularly that are doing the large language moral trainings that they need the capacity to do so.

And I'll pass it over to, Yeah, I, I think it's a question of scarcity breeds, uh, you know, people headlong rushing into these things and probably over provisioning, right? If, if you, you know, if you say, I, you know, I, I, you can buy these glasses, but I only have one pair and I've got 10 people who want it. I have some, I have some GPU capacity on a desktop gaming system at home that's I could rent out If anybody needs it.

That's right. That's right. I don't know what kind of GP those are or what GU stands for in your world, but I I don't think they're in video.

My steady application needs your GPU, so. Exactly. Hey, I, I think we've, we've talked, hey, but if anyone out there has any spare rooms, they wanna rent out it to become a data center, might be, might be a good time to think about it.

We're gonna take a break. We're coming back. John kind of, uh, telegraphed a little bit.

And video is the latest big tech company to under the watchful eye of our legal, of our government's anti-monopoly division. Let, let's see what it's about. You are watching Textron Gang.

We'll be right back. Discover Textron Group, the epicenter of tech innovation. We are your go-to for reaching IT leaders and practitioners worldwide.

Our secret impactful content that sparks awareness, engagement, and top quality leads with us. You'll access editorial websites, streaming videos, virtual events, custom content analyst research, and more. Join our satisfied clients.

Let's revolutionize your tech journey. Contact us today and tell your story to the world in the most powerful way with Textron Group. Hey, we're back with techron Gang and um, justice Department.

It's very busy these days. It just subpoenaed Nvidia, uh, late Tuesday night. And it's looking into collecting evidence or data about its market dominance, including its acquisition of run AI and its ability of customers to switch to non NVIDIA chips.

This is continuation of this ongoing investigation. And if they find enough evidence after talking to Nvidia and some of the companies it works with, it will probably then move on to an informal investigation. Now, on top of the Department of Justice's case with Nvidia, we have their recent victory in court against Google and federal court.

They are looking at Apple and I think the Apple store, the app store, the FTC. Meanwhile, it has sued Amazon along with 17 other states. And the FTC is now looking into OpenAI.

And what I want to ask all of you is, is this too much? Is this overexuberance on the part of the Department of Justice and FTC and what impact was this gonna have on the economy eventually, if they keep going after these companies that are so successful that are helping run our economy? Well, I wouldn't call it a, a witch hunt, I would say they're just kind of doing their job.

Does the job matter? Probably not because, um, you know, at least from what you can see so far run ai, the only thing AI is about run AI is the AI after its name. 'cause everything else is about operational software from there.

Um, there are alternatives to Nvidia out there in the open source community. It's still early days, and usually open source is trailing somebody who drives a PA pace of innovation. But we see that eventually come around.

And I think the market will drive that eventually. Um, yes, the, the lock in on Nvidia is not the hardware, it's in the software. It's in that Cuda library that they use.

Um, nominally that could be open source, but right now it seems like the only people who make any con contributions are all Nvidia employees. So maybe they need to kinda make that real open source and make it a community like Linux is a community. But I feel like it's a moment in time and I'm kind of okay with the DOJ at least sending a note that says, Hey, we're watching this and you should, you know, behave better.

But I'm not sure it's gonna actually amount to more than a mole hill. Yeah, yeah. No.

Can I just, I'll, I'll one more thing then I'll get outta the way. Um, I think aside from possibly fines, I don't think they're gonna be structural changes to these companies. At least I, I hope not in, in the case of Google, although I think DOJ in, in a sense was justified in the, that, in that action.

But, uh, it would be interesting to see what the punitive damages are in the, in the case of Google and in, in the cases of other cases that they pursue. Yeah, I said it before. I'll say it again.

I view Nvidia, Google, apple, Microsoft, Amazon as strategic assets of the United States of America. And I'm not gonna stand here and let this court right. Talk badly about our country.

Right? I'll, I'll take the Animal House Hoover approach. We'll walk outta here with Senator tki.

But, but seriously, um, you know what? Stop him. He is on a roll, you know?

Yeah. Don't, don't stop me with the Germans, Bob Pearl, Pearl Harbor, right? But, um, but, but seriously, ai, the battle for dominance around AI is gonna be the fight of the decade coming up.

He who controls, or they who control the AI are going to have an economic advantage in the world. We have. And as a country, we have all the cards, we have the winning hand here.

Why would you want to, Who, who is we Kimmo Sabi last time I checked businesses do not make decisions based on, You know, this, I'm not gonna respond to that kimmo sabi thing. I want you to go right now. But Businesses do not make decisions based on the strategic interests of the United States.

They are making business decisions Based on what's good for that business. And, and I count on them too, right? But, but having a strong tech industry that leads the world and is second to none in an era where tech rules is strategically important as strategically important as it was for the United States to control the oil industry, the energy means of energy production and stuff like this.

You know, when you look at the, the arc of world history, every great country, every, you want to call 'em Empire Nation, whatever you want did. So because they controlled some dominant means of whether it was transportation or manufacturing or in this computer age of computation. And where the United States of America, we invented the internet, we've controlled the internet.

It's a worldwide phenomenon. There are plenty of companies who look enviously at how we grow these companies right out there in Silicon Valley where John is. And Can I mention, can I mention something quickly?

Alan? Uh, and you're so, you're so right. When you look at the Obama administration, I remember covering him at least three or four times where he came out here.

He had the round famous round table where he's flanked by Zuckerberg. And I think Steve Jobs, uh, it continues, it, it started actually with Clinton in a certain sense. It, it accelerates through Obama.

And then ironically, uh, was it 2019, we, we start hearing about these cases from the Trump administration only for it to escalate through the Biden administration, which I find incredibly interesting. Yeah, no, I hear what you're saying and I agree with you, Alan, very much so. You know, these are national treasures, if you will.

The other piece of it though is they be, they're a national treasure because of competition. And, and I often say, you know, recently when I've been talking a lot about, you know, VMware or Red Hat, your Red Hat is dominating that Kubernetes space. We need more competition in that space.

So I'm rooting for Tanz to come up. The same thing that I'm doing here is we need more playing because when we have competition, you're 10% better, you're 15% better, you're running that much faster. And that makes Nvidia step up their game and not get, if anything you, you know, Intel took their eye off the boat.

They're now scrambling to get back there. Um, because they're threatened by a MD That's a wonderful thing. Yeah.

They're having to reinvent the company to say whatever. And I think the same thing is here now, is her motivation, is the motivation of the d you know, the, the right motivation of why they're doing this. Maybe, maybe not because they view competition or, uh, capitalism maybe differently than you and I do.

Um, and that, you know, they, these are all goodness for us, but not to where it's goodness for us to be able to make sure there's a balanced approach to the market. And there's not a single hold on it with any company. So I don't necessarily agree with it, but I think the action, it's too early to kind of start looking at Nvidia.

You know, come on, the market could crash tomorrow. Right? Anyway, I'll stop.

I'm gonna be the cynical one here. I'm gonna, I'm gonna join You. Let me talk ago.

Go ahead then. No, no. I, I think we're talking about the right things.

But, you know, calling a company a national treasure, first of all, I don't want Nicholas Cage showing up here trying to, trying god damn national treasure, treasure. So they, they aren't national treasures. What?

They are strategic assets. They're Strategic assets. They're Strategic assets.

I incorrect. It's not a national treasure, I'm sorry, Nvidia Jensen, you're just a strategic asset. That's all.

They're Strategic assets. We'll have to sell for that. Right?

And the question is, um, the things that we can do to help those, uh, strategic assets deliver what they can, what they, what they bring to the, our country and our economy and, and our, our social fabric. There's also the, and, but do things get outta control? And my question about Nvidia and, and the DOJ looking at that is what evidence do we have that they're doing some unfair practice?

Yes, they have 90% market share. Okay. You're gonna get a lot of attention when you have 90% market share and a massive, you know, CapEx or I mean, uh, capitalization, um, valuation of your company.

So it, it's gonna get scrutiny. And, and to your point, Kimberly, it's probably, it's too early as far as I know to say, Hey, something weird's going on there. Um, the constraint seems to be more on the manufacturing side and let's get those chip fabs into our country so we can produce more of those and not be rely reliant on that strategic asset that someone else has to the degree we are now.

So, but that, I don't think it's about protecting it, Alan, and don't, don't touch it and damage it. You know, it's a good thing and we'll, we'll just screw it up. Yeah, we'll screw it up if we overregulate it and, and start mucking with it too much.

But how do we get the most out of it and the most advantage that we can? I think when times are good, all these companies, whether they're car manufacturers or chip makers or whatever, they're all like, we're heavily into globalization. We're global entities.

We're making decisions for our shareholders, and then we'll benefit the, all these countries around the world. And then the minute they go broke, there're suddenly national treasures with a handout going, Hey, help me, help me, help me. The United States of America and taxpayers.

And, and what comes back to the taxpayer, exactly. Out of all of that is maybe how many jobs are in data centers? I don't know, but it's not nearly as many as a manufacturing plant and some other industries.

So I'm kinda wondering what's the return on investment of the United States taxpayer and these national treasures that we are investing in. So I would, the US isn't investing, We're bailing out intel. That's an investment.

Well, and and to caribou's point, right? About Intel taking their eye off the ball they did. And had they not, I mean, maybe this would be a different conversation we're having.

Yeah, That's interest. That's interesting you said that too, because remember when Microsoft took its eye off the ball around the internet? Google Yeah.

Filled The vacuum and then, and it opened up to Google and Apple got in there, turned That Up. Facebook. Yes.

But let's remember that Nvidia NVIDIA's not an overnight sensation. They've been at this for 20 years, right? It is a 20 year overnight success story, number one.

Number two, I'm not saying we need to bail them out. I say let the market have its way. You know what, I think Google took their eye off the ball on search.

I think Google has got so AI heavy on their search that their search doesn't, A lot of people, me included, when I search on Google, I don't get results that are, are useful to me. I get a lot of Google AI gobbly g**k, and I'd love to find a better, a better search right now. And OpenAI just started something called chat, search or chat, GPT search or something.

I forget the exact name of it, but I was reading an article yesterday where they're going to use AI for a Google-like search, but give you a different slant, a different search result, or a different way of looking at search results. I think that's the market. Let that happen.

May the best company win. But when the, when the government starts putting their hand on the scale, because you know, they wanna make it equal. Like we have some, and I, and I'm not bringing this to affirmative action, but that we wanna have some affirmative action for, for other companies to compete.

So we have a broader market that's not market economics. I don't think that ever works out as the government intends anyway. But that said, Of course not.

I'm Just, I would still still say on the other side of that equation, when the government makes funds available to these companies to keep them alive and 'cause they screwed up in this market that you just described, is that really something we should be doing? And if it is, then what are we getting out of it? Lemme ask you, did you agree on the, As the CHIP act?

And I think the CHIPS Act was really needed not free. Absolutely. Again, as a strategic asset.

And it wasn't. I mean, yes, Intel has been the biggest beneficiary because they, they kind of pivoted around this chip act, but the, the company, what is it? Taiwan Semiconductor, whatever, Micron.

Um, They're all, they've all gotten money. 'cause we wanna build chip foundries here. And the, and that's, I think that's a worthy, worthy thing.

But you know, this takes you back to the recession at the Great Recession and the auto company bailout. Mm-Hmm. Did the u did the US consumers get their money back with their interest?

Yes. Did we get a stronger auto industry as a result of it? That's still US based.

Yes. I don't, I don't think that was a bad deal. What are we getting out of this chip act again?

What's the return back to the taxpayer? I, I hope we're going to get high paying jobs at these chip foundries. And more importantly, we, our entire economy runs on CPUs and maybe GPUs soon.

And with you, you have a strategic, not adversary, but a a strategic competitor who decides they're gonna rise, raise the price of those or cut off your supply. It affects all of us. You can't drive your car.

There's, there's how many CPUs in the average car today? It's Also a national security issue, right? Invade Taiwan, you're screwed for chips.

Black Fellow and, and, and, and, and how did we get to this situation? Oh yeah, globalization. We're gonna push all this stuff overseas and manufacture there and what a great idea that was by a bunch of MBAs in Wall Street, most likely.

And now here we are screwed the other way. 'cause to your point, we won't have semiconductors. So it is a national defense issue.

But you know, this great market thing that you described got maybe a little too far ahead of itself and we didn't think through these strategic issues because, well, that was the best thing to do on Wall Street and let the market decide. But that's, that's our System. Yeah.

Kimberly, make sure, sorry, It's ly These lids getting exciting. It's exciting. Well, I mean, one, uh, I was a little, probably about six months ago I held, had a panel on the chip sax and we bringing a bunch of people that are, have been involved with some of the decisions that are going on there and started looking at where this money is going.

And one of the things that's wonderful about this is we're seeing middle America, they're dropping in lighting up America partially because they're a little bit easier to get regulations through to build the facilities. I mean, you look at what's, you know, Phoenix is blowing up with this, uh, we see it in Texas. I'm not sure if it's going down in Florida.

We probably didn't put anything. No, no. Nothing goes down in Florida.

Even New York state for crying out loud, you've seen some major investments in New York State up in Albany with IBM, um, and Intel. So there's, you know, this is spreading us across the United States that is going to benefit the kids coming through schools. You know, because now I'm coming through, you know, now you're gonna see people like not only just the, not just the, the, the coastlines, but are the middle of the country having tech in their backyard.

And that raises all boats. I I like that idea. But Mi Mitchell was next.

All I, I'm really glad you, you're really glad you shared that Kimberly giving some real insights to what the impact of the chip act. And that's, it isn't capitalism and globalism bad or regulation is always bad. Yes.

All those things can, can run amuck, right? And, and that's our system of, I wouldn't call it checks and balances, but it's, things get too far outta whack. And some, some other counter reaction is necessary, right?

So the chips got outta whack, we sent too much overseas and now we're vulnerable. Okay, great. Make an investment.

But I think to your point, they did it in a wise way. 'cause it's an investment that also grows economy adds jobs and provides national security. Now we could have just, you know, thrown a bunch of money, which we've done during Covid, thrown a bunch of money at companies and say, here you go spend it.

And then they all do stock buybacks. That's stupid. You know, that kind of stuff doesn't help us.

But this is just, I think it's a natural part of our system. Um, corporations are gonna be driven by shareholder value return to shareholders. And that doesn't always operate in the interest of other people.

I mean, yes, it's competitive, but sometimes it goes a little too far and we gotta pull back or we have to give others a fair chance to play in that market too. These add job things. I'm not so much biased.

The chip manufacturing plant is highly automated and the data center is getting more and more automated. There's a standing joke that says the future of the data center is one guy and a dog, and the dog is there to keep from touching anything. So these are not manufacturing kind of plant jobs.

So yes, there may be a handful of high paying jobs, I'll buy the defense argument, but I'm not buying that. These are creating jobs in middle America argument. 'cause I think they're small.

Well, I don't have numbers at my fingertips, but they are significant. I, I think they are too. But I also think that's the role of government, right?

To, to stimulate economic development. And you know, uh, like we have hub zones, right? Uh, where, where, you know, minority owned businesses get breaks and tax incentives to bring business to and bring jobs to under, under economically developed areas that, that is a valid role of government.

But look, it is what it is. I I, I wanna wrap a bow on this. It, it's too early to be slapping NVIDIA's wrist and try to inhibit their growth here.

I, I, you know, let the market rule. I, Kimberly you mentioned, we found we could use CPUs instead of GPUs for a lot of things. Let let the market innovations fly here and then come back to us.

But who knew we were gonna to get a lesson on socialism, Marxism, capitalism, economic theory, all, all out of here on text drawing gag. We'll take a break. We're gonna come back and we're gonna tackle some, is it physics and calculus next?

I don't know, maybe it's the future of meetings in ai. You're watching Text on Gang. Welcome back to Textron Gang.

I'm Jon Swartz. Hey, do meetings, could they be productive and even enjoyable? I don't think so.

But software company Atlassian does. And last week they agreed to buy rewatch a platform that records summarizes and organizes meetings and screen recordings via ai. The the plan is to meld rewatch with Loom, the asynchronous video messaging platform that Atlassian acquired for nearly a billion dollars late last year.

So basically, in addition to that, Atlassian has grand designs for rewatch in addition to integrating it with Loo. And it tends to fold rewatch into its vo AI platform, which would enable meeting notes to become j shoes transcripts could also be searched within the business context. I guess the idea is to make meetings more productive and enjoyable since that's what we seem to do all day.

I don't know what you all think, but at least this is a intriguing idea. Something that might have some legs to it. I don't know what y'all think.

Um, let me add some context to this. Um, I met with Atlassian during Team 24, uh, when they introduced roe, um, and RO o So, so yes, this action of, of, of, uh, buying the company, um, rewatch. And they also bought Loom for video.

But what that fits into, what what Atlassians has done has created a, a platform for AI that under underpinning VO that actually supports all of their applications that, you know, they have many applications, whether it's for, uh, ITSM or for development, et cetera. Um, and they bought Loom because they needed a way to communicate other than just through text and messages and, and tickets and things like that to add some rich media to it. So what's driving that is not only an AI platform and by, by platform I meaning it's not 20 different AI things happening in 20 different products.

It's an AI platform that can be leveraged by any one of their products in whatever ways that makes sense. So rewatch to me is another component that can, can access and use that ai. And what's driving the AI isn't just machine language and, and, uh, expert models and, and gen ai, they, they add what they call a context or a team graph.

So they know I'm Mitch, here's my user, here's all the things I do across the ecosystem of Atlassian products and the data that I use, the data that I create, the data that I need. And the potential of this is how much of us walk away from a meeting saying, okay, that was over. I took a few notes, I forgot where I put 'em.

And nothing comes out of it, right? If you have some, not only a recording, but if you have some assets that AI turns it into some meaningful content that can be shared with others and maybe you shared and used by you in some future workflow. That I think is the potential of, of looking at why they're using Loom and why they're using rewatch is they're really working on workflow and how they can leverage AI and, and data that, and then the context of which you operate.

And, and I think that's the, the secret sauce that they're building. I think that this is a lovely way to compensate for the fact that most of the meetings are run by people who shouldn't be running the meeting. And basically what happens is they don't take the notes, they don't make the assignments, they don't remember who said what and who's responsible for what.

So, um, it will be nice to have a machine to do that, but what I really, really, really want is an AI to evaluate the purpose of the meeting in the first place. And maybe say, does this discussion warrant a meeting? Because half the meetings you go to were probably pointless in the first place because either their standing meetings that somebody feels that we ought to have every week just because, well, they're on the schedule, or B, somebody decided to call a meeting because well, they couldn't figure out what was going on.

So they called the meeting and instead of just, you know, calling the people who are involved in that workflow and say, where are we? And so instead we'll just put 12 people in a room and inflict this pain in common together. So, so enough to solve Thank you.

From the man who doesn't log onto to Slack. Yeah, Exactly. Um, There we go.

So I feel like this is like a genuine, ignorant kind of thing going on today. But, um, Anyway, I just picked that one up. So, okay.

Reference, my husband will get along together. He uses that phrase. Oh wow.

So Mary, here's my take on this. With all due respect, meanings serve multiple purposes, especially, and we saw this in Covid in spades and it, and it still remains 'cause we have so many people who are remote, much like here. And Textron again, the banter between Mike and I is very different than the banter between our three west of the Mississippians.

And it's not because you guys live west of the Mississippi, it's because we're on this medium. No. Whether it's Zoom or Streamy yard or whatever you are on, it's not the same as an in-person meeting.

So you do need, I think for all the things that John and Mitchell and and Kimberly have laid out, you do need someone to take the notes, summarize it, give the action items, and it would be good to have a person do that. But I, I frankly think it's a better use for an ai. An AI should be able to do that really well.

I I disagree on that one point. The banner is good because What a surprise Because we're both New Yorkers and we're born to argue, right? So that's why we do Yes, and that is true.

We, we, we do that. We do it well. But, and, and to that point, there's a cultural aspect to doing these meetings.

There's a bonding that goes on, whether it's in person or, or in this kind of medium. There's a cultural aspect to it. And you know, and that give and take and that banter makes, makes for better relationships.

It could be laughing, it could be not laughing, but so to say, wow, I'd rather not go to the meeting. 'cause I think they're a waste of time because we do it every week and I'll just, I'll, I'll, I'll, you know, I'll do the the cliff notes, I'll do the AI notes. I'm not against the meetings.

I am against too many meetings that don't have a purpose. And so there's just not enough. The bar isn't high enough for what the content of the meeting is, and I'm all for the social aspect of that.

But there needs to be a purpose here for the meeting itself, beyond the fact that you and I wanna Just talk Yankee baseball. Yeah, exactly Right. Um, We end frustrating things for me is, is, and so I'm gonna be doing three meetings back to back to back one on, um, me teams, which I like.

And then the other two are on Zoom, which I don't like. And one of the frustrations I have as a reporter at least is you want to have some sort of eye to eye contact with the person you're talking to, but it's really difficult to take notes when you're, you're looking at somebody through the camera. And so consequently that, that's one of the things I don't like about it.

I normally talk to the person on, on camera for a few minutes and then I, then I shut it, shut it down and start taking notes on my computer. Um, there, there was, I did a meeting of recently where we had an AI agent of some sort take notes and summarize, which was actually pretty good. So if, if, if Atlassian can get to the, any, any further than that, I'm all for it.

You know, I I'm looking for any type of useful product tool To so to a meeting. Yeah. Maybe, um, I mean Mi Mitch, Mitch, she might have the answer to this.

I mean, we have Otter AI that's out there. I've seen other attachments to Zoom to do this kind of work. What is you unique or different about this that we should take away because we're seeing this implement?

I mean, my peers use it to take notes. I've, I've had, um, others, you know, kind of email me the summary of, of the meetings and that sort of thing. Uh, the difference is, uh, Atlassian is not a meeting platform.

They're not team, they're not Zoom, they're not stream art. They're not, you know, uh, Google Meet. They are a recording and an AI analysis of of that recording, right.

In whatever context. So maybe to Mike's point, and this is not what they've said, but they, they don't require that you use a Atlassian whatever meeting software 'cause they don't have any. Um, but they are, they are adding Loom into more of their product.

But apparently, as the way I understand it is they'll still be able to, to work with, uh, with Google Meet and Zoom and everybody else's third party software. Okay. Did I get that right?

So I think it's then what it is Kimberly is taking what happened in that and then putting in the context of the people that were in it, that graph content and graph. 'cause I know about the five of us and I know the kind of work that we're doing across the applications that we use within Atlassian and being able to add some context to it, of, of the five action items. Seems like two of those were in Mitch's camp because of what they are, because I know what he works on and what part of the, the ecosystem of creating software or running the data center, he, he runs on.

So I think that's the real value of this is we all have all kinds of Otter and other things that we can add to our meeting software. It's now how do you take the output of that and use it? And that's what this is really about.

So, so Mitch, back in our still secure days, we would call that actionable intelligence. Yes, exactly right. Odd as a recorder.

Rewatch can be a recorder, a summarizer, whatever you want. You don't have to have, uh, a Zoom platform to have an TER or whatever it works with whatever plat it should work, you know, platform, uh, agnostic. But taking that information and not just recording it or bullet pointing it, but putting it into actionable intelligence.

Got it. Right. That that's money.

And I think that's what Atlassian is, is going for here. I I agree. I Agree on actionable intelligence and after that, nothing else.

Okay, here's the problem. The garbage in, garbage out. So we assume there's, in that, Here's the thing that I do like about what Atlassian theoretically can do, right?

So they sell the project management software that people use to organize their software development or whatever else it may be. Um, we use some of their software here and it's, it's awesome. The part that would drive me crazy and still drives me crazy is, so let me get this straight.

We have a project management application. Let us call a meeting of everybody so I can collect the data from you so I can enter it into this project management application. When what we really wanna have happen is for that project management application to be automatically directly integrated into your DevOps workflow or whatever else it is you're working on.

And it just automatically updates that project management application, thereby eliminating the need for the meeting, eliminating the need to record the meeting. 'cause the meeting was unnecessary. That's what they're trying to do.

That's what they're going after. Mike, is instead of having silos of tools and data that those those tools use is the only way to effectively use that information. 'cause you don't understand in what context is it useful outside of that event or outside of that tool.

And that's where the workflows and content graphs and things like that make sense. So how, you know, whether they gonna solve the world's problems, I don't know. But they have a big enough ecosystem of applications that they can do a lot of things.

And I think they're on the right track in many, many ways. And I think potentially to the point they may be a model of for others to emulate or at least learn from and start to do some of the same things. You can ask, why isn't Microsoft doing this?

They're, they're, they've been in this position for a long time. They've got all these applications that we all use and, but you know what? They all operate independently and just show up on SharePoint.

Who cares? 'cause the government slaps their hand when they try to do it. Oh, okay.

Well there, I I would come back to this meeting thing for a second though. So what I'm hoping out of all You can off with a bone over here. Huh?

What what I am hoping out of all of this is that when that data's already collected, we won't spend half of the meeting just trying to figure out what the data is. We can actually talk about the thing that matters and the thing that we're trying to get to the outcome of, and that makes the meaning valuable. I think there's a ver bizarre persona at, uh, Atlassian that they're using to help design this Thing.

I if they're not, they should. But anyway, Hey, I gotta pull the plug on this meeting before Mike Overboard. Um, what a, what an interesting and fun Textron gag this was.

I hope you guys watching this had as much fun as we did recording it. I'll remind you we also have it. Uh, speaking of fun, we have an added feature today.

We've got one of Ira Winkler's bite me. So, uh, solos coming up. So in Ira, IRA Winkler's bite in the episode, uh, for, for today, IRA takes a stand against, you know, keyboard warriors who, who are all about showing support and expressing their opinions from the an anon anonymity of sitting behind the keyboard there and, and saying it.

But yet their actions don't match their walk in real life in terms of morals and ethics and, and stuff like this. And it, and it's, you know, it is, we, we have a holier than thou social media, you know, echo zone. But people don't really follow that mantra in real life.

So IRA's always funny with these things and, but biting. So here's Ira, IRA Winkler and bite me. Hi, this is Ira Winkler and on this edition of Bite Me, I want to talk about ethics and what it actually means.

And frankly, this is kind of related to the whole black hat women lamb shaped things, which again, is so horrendous to begin with. But what bothers me most about it is all of a sudden people started showing up with what I call keyboard courage. And if somebody took a picture, and I guess it's good they took the picture, but did nobody ever go ahead and decide as they walk by to stop and say, Hey, this is royally f****d up.

You know, again, I'm cursing, but literally it's royally f****d up and nobody ever went and approached them and say, Hey, don't you think you should get this down? I don't care who thought about it, and I'm not talking about the people who worked there. I heard, I'll go into this in a second, but you know, like the people who worked there might have been too scared to point out that they're wasting a lot of money on something that's incredibly heinous.

But how many people walked by the thing? I mean, I'm assuming it was like a little event in a corner or something to that effect at some side event, which I, it was, but the reality was how many people walked into the event and didn't say anything? How many people walked by it and took a picture to put up on Facebook later instead of actually saying something?

The issue is, and again, why I say this is ethics as opposed to anything else. There's a phrase I heard a long time ago that ethics is what you do when nobody's looking. Ethics is not something, you know, ethics is what happens if you are in a situation that you don't think you'll get caught.

That's number one. But ethics is not being, okay, let me put this up and let me start making a statement on LinkedIn after the fact. Ethics is doing something when you can do it.

Now, it's great that people started saying that this was bad at after the fact to a certain level, but it almost became the obligatory badge of honor. Let me join in, because let's face it, everybody knew this was f****d up from the beginning. You didn't have to go ahead and say, I personally agree it was f****d up.

That was I, A lot of people are just doing that for likes, in my opinion, and you know that's true. It is f****d up and they need to say that. But more important, you need to say it when it's going on.

You need to say it when there's something that could be done about it. Now, in this case, they're saying, oh, this proves, you know, the environment is hostile to women. I don't agree.

And the reason I don't agree is because yes, something will be royally f****d up for women, especially women are abused constantly, and I don't want to say in throughout our industry, they are put down, they're snide comments and things like that. But ethics goes into, and again, this is why ethics are important the way I'm phrasing it. Do you do something when you witness it?

Everybody all of a sudden was saying, oh, we heard that. You know, I always heard that inside that Palo Alto was not friendly to women. I'm like, well, why didn't you say something?

Not until after this when there's no way to prove it. It's like, well, somebody told me behind the scenes. No, you know, it's not the time for all of a sudden people come out.

I knew about it all the time, but wink, wink, that tells me you're a f*****g a*****e without any ethics to begin with. Because if you see it and you do nothing about it, you're complicit with it. If you see it and do nothing about it, until you could take a picture and put it up later, that's kind of, doesn't make you the person who acts.

It makes you kind of the person who wants to go ahead and have keyboard courage. If you really wanna support women, stop having keyboard coverage, keyboard courage, and have courage when it matters. Have courage to stand up, walk out.

If you don't like the organization, you don't accept things from them Later, you go ahead and you take the, you know, I got into a discussion. It's like, well, why should I be punished when other people are doing bad things? You're not being punished.

You're taking a stand. And the point is, if you really want to have ethics and you want to take a stand, take a stand when something could be done about it. 'cause I promise you, hundreds of people walked by those women with lampshades and could have taken a stand and done something about it and walked up.

And if the environment was hostile to women, then having other people come out and say, this is royally f****d up. We'll go ahead and have an immediate impact, not keyboard courage later, which gets an apology, and rightfully so. Maybe they don't realize, I don't know.

I'm not gonna try to justify this, but you need to take the stand when something could be done about it. Not have keyboard courage after the fact. Anyway, this is Ira Winkler with this edition of Bite Meat.

I was right and I would have a meeting to help reform meetings, but the problem is that's counterproductive. So I'm kind of stuck in a Quandary. Alright, I'm, I'm referring Mike to the department of Take a Stand, Mike, take Monopoly at meetings.

Anyway, guys, what a great, what a great episode this has been. I appreciate you all. Kimberly, John Mitchell, Mike joining in with me here today.

Appreciate all of you out here watching this. I hope you enjoyed it. Again, we have a full lineup of text show TV behind this though, and I hope you'll watch that as well.

This some really great interviews and sessions and great information there. Podcasts to check out. We'll be back tomorrow for our Friday end of the week, and if, if this was any indication, it's gonna be a, a heck of an end of the week show.

But until then, is Alan Shimel. On behalf of Techstrong and Futural Group, thanks for joining us. You've just watched Textron gag.