The IPO Comeback: Why Tech Giants Are Finally Going Public | All-In Liquidity IPO Panel

00:00:00Hey, 2026 could be an all-time [music]record for IPOs.>> The AI IPO of the year so far. Thatcompany is Cerebras.>> Cerebras Systems founder and CEO AndrewBubman. We are participating [music] insomething extraordinary on everything wedo. We are the fastest bar none.>> Well, Marshall is the co-founder and CEOof Planet [music] Labs.>> Space and AI are really um a match madein heaven. They're getting married. Infact, just like Google figured out howto index the internet and make itsearchable, we are indexing the earthand making it searchable. [music] He'sgot his glasses, the famous red glasses.Brad Gersner is here, founder and CEO ofAutoimmet [music] Capital, a leadingtech investment firm.>> I believe that the wave is the biggestwave in the history of technology,[music] will be incredibly beneficialfor America. I'm rooting for all of thembecause I'm rooting for America. Ladiesand gentlemen, please welcome BradGersonner, Will Marshall, and AndrewFeldman.

00:01:00[music]>> On the couch, we switch on the couch.>> Nice to see you, my friend.>> Hey, [music] big boy.>> Nice to see you.>> Last last time I saw you, we were inDavos.>> Yes,>> we were in Davos causing another drop.Another JL. Do you hear that littleDavos?>> We were just, you know, it was preipo.We're chopping it up>> with Davos.>> We're in Davos.>> Hanging out at Davos.>> Well, no. Listen, I was Everybody knowsthe story. I'm supposed to go on myyearly Japan ski trip. Sax calls me>> with Tucker.>> Yeah. Well, anyway, we don't drop thatname, but [laughter] I'll pick it up foryou. Put it over here.>> Tucker. Anyway, so I cancel on Tucker. Icancel because Sax calls me. He says,"Listen, Pus needs you, the world'sgreatest moderator in Davos." I said,"No problem.">> I said, "Saxs, POTUS, and Davos." So Isaid, "When?" He says, "In 3 days." Isay, "You got it." I go and they give mea badge. And it's like the special greenbadge and they buzz you through thesecurity and I look at the monitor andit says Jason McCabe Calacanis withDonald J. Trump.>> Oh wow. How did you feel?>> I thought it was hilarious. [laughter]So then I went and we did a greatinterview there and we did like six orseven of these great

00:02:15all-in interviewsand it was fun.>> Let's start this because uh the two ofyou guys run two of the most interestingand consequential newly public companiesin the stock market. Andrew Feland isthe founder and CEO of Cerebrus. WillMarshall is the founder and CEO ofPlanet Labs. But you are also theinsight and a gateway for all of us tounderstand these two big trends. One isin AI silicon, the other one is in spacedata centers. I think it would be areally interesting thing to>> and emerging.>> And emerging. Yeah. Um but let's justtake one step back. Uh you just heardthe last conversation about beingpublic, going public early. Let's justtalk about that cuz I'm just verycurious. How's it been? It's been 3weeks or so for you. It's been about ayear and a half or two years for you. Uh>> it's more fresh.>> Was it Was it everything that youthought it would be? Like>> what's clear so far is I need to upgrademy namerop game. I mean that that was atour to force. [laughter]>> But by the way, you were you were inDavos with>> J. I was I was there but that um uh>> tour [laughter] to forceuh look I I I think you do all this workand I I think it's really difficult toto overestimate the amount of garbagethat's involved in in going

00:03:30public. thenumber of meetings where you you look onthe the Zoom and there are 130 attendeesand the amount of times you review thesedocuments and the commas move and andjust no value added. You go there andyou have this enormous event and thenext morning you've sold no more stuff.your engineering projects have made noprogress since the day you weren'tpublic. And you go back to work and umyou you have some new constituents thatthat you have to to to address andcommunicate with, but the core parts ofyour business, you have more money inthe bank. Um but not a damn thingchanges in the important parts of yourbusiness. um if you still if you neednew supply or if your relationships withyour vendors are bad, they're still bad.If they're good, they're still good. Andand so I I think what what we've seen isum you your employees have a party,everybody's really excited, and you putyour head back down, you high five, andyou go back to work. Can>> can I can I just give a little contextand then I want to hear from Will. Youknow, if if I can, Andrew, you know, wewere investors

00:04:46in Cerebrus. I was on theboard a year earlier where we weretrying to go public. Um, and you know,aside from just being a warrior whoweathered a decade worth of storms thatwould have taken out any normal humanbeing, the path to going public forCerebras was a particularly challengingone. One of their investors was the UAE.So there was questions about CPHAS, youknow, in in the prior under the Bidenadministration challenging to getpublic. My observation outside lookingin is everything was really hard untilit got really easy like n 9 and a halfyears of really hard and then 12 monthsyou know of of of really easy whereeverybody wanted to get in.>> They priced the IPO at 185 which was upthe range was taken up two times. Okay.The stock opened at $320 a share Ithink. Today it's at 230 bucks a share,5060 billion dollars in market cap for abusiness like you know and andrew isjust one of these people. Let's let'sget back to work and build But mymy just add-on question to that is froman employee morale perspective likedistraction

00:06:01perspective etc. has thelast 3 weeks you got a lot more capitalyou got a lot more profile presumablyit's easier to sell to enterprisecustomers today. net net if you wereadvising me if I was in a similarposition would you say go public?>> I I I think the first thing is a lot ofpeople asked us about how we got thetiming right.>> Right.>> And I I think the answer is by gettingit wrong for a decade. I mean that'sreally the right way to get timingright. Um I I think um first I we wewe've been at this for for more than adecade and and we we brought everybodywho'd been with the company more thannine years to share and we brought theirfamilies. [snorts] And first I learnedthat engineers owned ties. I didn'tactually know that. [laughter] Um andthey didn't die when they wore them. Andsecond I was surprised at how big a dealit was for them and their family. theythey were really proud in a way thatsort of their parents [clears throat]might have heard of it or uh that thatsomehow this was like a a bar mitzvah orkinera or something.>> Um>> and then you you had these sort of thethe children of immigrants, one of ouruh one of our leaders, her

00:07:16father,Chinese immigrants said, "I thought itwould have happened faster."[laughter]>> Right. And but I I think um we are sortof by natureuhsort of in the uh in the trenchespeople. And so um we we love solvinghard problems. And so when when we hadthis excitement, everybody went and theywere so excited and we had a party andum I I think it it gave externalvalidation. And then everybody turnedaround and said, "Now what are we nowback to work?" And and so I I I think um>> and so you you started off kind of bangright out of the gates. Will you had alittle bit different experience in termsof you know the entry to the publicmarkets, but over the last 12 monthsyour stock has gone from five bucks ashare to 50 bucks a share, some 10x movein the public markets. So talk usthrough the other side of this where youcome public, nobody really notices untilthey notice.>> Well, we were one of the first spacestocks and and I think people just hadno idea what earth is going on in space,how it was changing everything and theywere just like

00:08:31what the heck is that?And and but but you know, I have similaropinions. I mean, in the end, you'vejust got to get on with executing thebusiness. uh the going public gives youaccess to liquidity uh for earlyshareholders whether that's the employearly employees or early investors andthat's great it gives you cash for thecompany that's great and I do think umit helps your business as well becausethe maturing event gives you morecredibility to various customers and forus we work with biggest agriculturalcustomers big governments civilgovernments defense and intelligence allof those sort of actors they want toknow you're going to be around.>> Exactly.>> And not going to disappear. I mean, wehave countries that are fully dependenton us giving them information. Theydon't want to just disappear. So, theyreally care that we're going to bearound and and being a public companygives you the kind of force in the worldthat people go, "Okay, you're here tostay and you have access to capital ifyou need it and so on, right? It'slegitimizing>> and you know, you you know where thestock is at any one day. You know, we'renot focused on that dayto-day. We'refocused on how we build long-term valuefor our shareholders, right? And um youknow the market is I think started toreally understand where space is going,why it is changing the the the world.

00:09:47You know people forget how space is partof your everyday life. Every time youuse a phone you're using communicationsusing satellites or GPS using satellitesor satellite data in some way or anotherthat's sort of integrated in your lives.you may not realize it um but it's justbooming now there's there's a>> and the the story's changed as wellobviously with SpaceX going public buthas the framing ofplanet gone from like a data source forpeople who need data from space and mapsto hey this is a tool to accomplishtasks and military like post Andrew'ssuccess like you probably would havebeen bucketed into Andrew as a militarytech company. So is that framing what'sdriving a lot of>> I think it's a bit more nuance thanthat. I mean firstly for the audience'sbenefit what planet does we havesatellites doing earth imaging. We havethe largest earth imaging fleet about200 satellites. They image the entireearth every day. So think of it like theGoogle uh satellite lay on Google mapsthat you can look at except it's today'sdate rather than 3 years old and we haveevery day going back. So it's a timeseries analysis of everything going onon the earth. That's useful for farmers.It's useful for energy companies. It'suseful for

00:11:02civil governments, floodingand fires. It's useful for securityapplications like you're getting at. Andit's a wide variety of use cases. Um, Ithink that where we're seeing this isthat AI is now enabling the it'sbasically reducing the barrier to entryso that more people can get access tothis, right? And uh, you know, there's alot more to say on that, but AI is onlyas good as the data it's trained upon.percentage is military. I'm curious.Sorry.>> What percentage of revenue customer basum uh security is part of the initialthing that we said we would do um out ofthe gate, but it's true there's a biggerfraction today than perhaps we wouldhave guessed. Um but the needs of thegeopolitical situation right now demandwhat we're doing. Um you know, just asan example, what this does is enablethem to see threats around the corner.>> Yes. and then uh you know give themweeks or months advanced warning ofthings and then that enables them tomore likely do things that stopconflict. So we believe this is you knowreally better for the world.>> Are you reticent to be perceived as amilitary company?>> Not really. I but I wouldn't say we'relimited to being perceived like that,right? um we we are helping farmers, weare helping

00:12:17uh you know uh energycompanies, civil governments, we workwith NASA, we work with what have youand and so um it's a it's a bigger it'sa bigger play than that but back to thespace uh piece of it what has changedobviously rocket costs have come downabout four or 5x over the last uh 10years which has helped tremendously buta thing that people don't know that isactually perhaps more important is thatwe've had a miniaturization ofsatellites. So that the same satellitethat used to cost a billion dollars andweigh 20 tons now cost a few kg or a fewtens or hundreds of kg>> and can do just as much stuff if notmore. It's it's the same as the sort ofmainframe computer to desktop revolcomputer revolution for space and it'sunlocking just like mainframes todesktops unlock loads of applications.This is unlocking loads of applicationsand it's so both go in combo the launchcost coming down and this>> let's build let's build on this. So Ithink I I'd like first you maybe take afew minutes and then I I want to talk toAndrew the same question. Both of youguys are at the foot of what areprobably huge secular trends intechnology. How I would frame this is weare rebuilding

00:13:33the data processinginfrastructure that has existed on theearth in the sky.And first you do the satellites, but Iwould love for you to explainspace-based data centers because I thinkeverybody's hearing about that.>> Are they really viable? What are they?How will they work? etc. And thenAndrew, this is the rebirth of silicon.We're going to find the next version ofMoore's law, which I think is moretimebounded, not transistor densitybounded. We now hear a lot about domainspecific architectures. we hear I meanyour chip was just a completetransformation in terms of the designprinciples that you know like at Grockwe took a very different approachNvidia's taken a very different approachyou took a big pizza shaped die and said it yolo this is it and you wereright just explain where we're going insilicon so maybe will you start and thenAndrew you start>> I mean what we're seeing firstly inspace is is all these new applicationsbased on data and AI so you know wewe're collecting vastly more data aboutthe planet. And with SpaceX and Starlinkand One Webb, they're they'retransporting far more data around theplanet. As you say, we're sort ofchanging um uh the nature of data usingsatellites. And that's basically doing

00:14:49what was once the province ofgovernments only and giving everyoneelse access to uh satellitecapabilities. And that's going to I Imean I I estimate there's a 75 to$100billion market just on Earthobservation. this kind of data wecollect and AI on top of that unleashingall that application. So that's thenear-term thing. Applying large languagemodels to earth imagery data, unlockingagriculture, you know, um energy, civilgovernment applications, permitting, youname it. This is going to makeeverything more efficient. Um and thenwhere we're going is indeed space is iswe did a study with uh our partners atGoogle about eight or nine years ago uhlooking at what are the costs of datacenters on the ground, what are thecosts that it would take to put them inspace and when might it make sense to uhdo it non-aterrestrially.And we figured out that when launchcosts come down to about $200 to $300 akilogram um uh it would be cheaper, justsimply cheaper to put the data centersin space. Now we're about $1,000 akilogram, just over that in right today.Uh but that's come down about 10x in thelast 10 years. Um on the currenttrajectory with Starship in particular,I would expect it the launch

00:16:04cost comedown there in 2 to 3 years. Elon mightsay it's next week, but at leastrealistically a couple of years. Sowe're not far away from it literallyjust being cheaper. Then in addition andthe intuition there that is that helpspeople understand that is you wouldnaturally use solar panels for doing uhthe the data centers are a powerproblem. It's a power game and you wouldnormally use solar panels. That's thecheapest way to get a watt uh today byfar. But you you don't want intermittentpower. So then you have to havebatteries or you then you have to havegas or then you have to have nuclear andthen it gets really expensive. Um inspace you can put a solar panel in asuns synchronous uh dawn dusk orbitwhere you're 24/7 looking at the sun. Soyou can have a solar panel that collectsand gathers five times more energy persolar panel than on the ground and youdon't have to have batteries or anythingelse. Uh so the infrastructure forcomput in space is literally just solarpanels and the chips and then the RFsignals up and down. So it's actuallyreally quite simple. It was just aquestion of when it's going to becheaper to launch all those solar panelsand chips into space than putting on theground. And it turns out that's going tobe in a few years. So we're

00:17:19partneringuh with Google to launch some of theirTPUs into space. We've already launchedNvidia's uh uh some of Nvidia's GPUsinto space. We're launching Google'sTPUs into space on an early test.There's lots of technology to figureout.>> Let's have a conversation. Um, but it'san early it's early days, but I think noquestion within 10 years most computewill be putting in space, which to giveyou a sense is a lot of money,[laughter]like trillions, and um will be biggerthan any of the other space businessestoday. comes Earth imaging. This is whywe're getting into this.>> Do you believe this? Do you believesending data centers to space makes moresense or is it just the regular>> Can you have him explain the businessfirst and then>> Oh, yeah, of course. Yeah. So, I thinkthey're, you know, with all due respect,one or two hard problems still left bebeyond putting putting GPUs in in inspace right now. I I think um we wewe're not super good yet atbuilding the clusters in space necessaryfor the communication>> between>> between Exactly. between>> We're not good at doing it

00:18:34on theground.>> We're not good at doing it on theground. We're really not good at doingit in space. I I think this is anextraordinarily important andinteresting problem and one we should bespending money and attacking. I've gotit in a slightly different time frame,but one that certainly will occur. Andthe the hard part is is it is it one ofthose problems where uh the last 10% is80% of the time.>> Now, self-driving was a problem likethat, right? Where the last 10% provedto be a decade's worth of work and justnow we're over the hump. And we don'tknow yet, but I think the interestingwork they're doing uh at Planet isreally important. And I think thefundamental driver to experiment to evenget insight into whether I'm right ornot is to get down the cost of launchvehicles. Then you can start doingexperiments and getting it wrong andfixing it and figuring it out. And untilthen it was unpaid.>> For the foreseeable future, you're goingto be terrestrial. explain your businessand how you made these criticaldecisions that kind of took you on adifferent path and you know you versusNvidia versus AMD and what you think thefuture of AI silicon looks like. I I I Ithink there were two parts your yourfirst question was around sort of therise of silicon in general

00:19:49and I I thinkwhat AI did and it's it's rarely sort offramed this way but it allowed computersto address a class of problems thatbefore AI computers were bad at.We were bad at images for almost theentire history of compute. We couldstore them and that's about it. We werebad at language. We could store it butthat's about it. We could transformnumbers. We were magical with numbers.And what AI did starting in about 201516 is it opened the door the aperture tosay maybe we could use computers onimages.All right. Maybe we could find insightin images. Maybe not only could we storelanguage but we could generate it. Allright. maybe we could understand itrather than storing it and regurgitatingit. And what this did is it it opened upsort of to computehuge areas that were previouslyforeclosed and at the same time we wereadding to those areas. we were takingvastly more images,all right, terrestriallyin satellites. And what this did is itit simultaneously opened

00:21:04up this entirearea and allowed compute to attack it.And this is what's underpinning bothNvidia's growth and and sort of all thegrowth you're hearing about in in AIcompute is as a as a processor builderas a hardware builder suddenly our toolscould attack more and different parts ofknowledgeand and that was sort of the first partto to to answer your question. Now howyou do that there are lots of differentstrategies tons of different ways toskin cats. What we saw in 2015 wereseveral things. First, we saw that AIwould be an enormous consumer ofcompute.All right. And historically for computerarchitects,new workloads were the opportunity forshare to change,right? Share changed when the rise ofgraphics emerged and you got thededicated GPU. That's how Nvidia wasborn. Share changed when cell phonecompute emerged and Intel and AMD whohad fabs and the best architects gotzero share and it all moved to ARM.Right? Share changed in the late '9swhen Nortell and all these companies

00:22:19we've forgotten about couldn't buildchips and couldn't do uh uh datanetworking and and what you got withCisco and Juniper and Arist and thiscollection of new companies. So we knewthat um this new problemwould present an opportunity for massivechange.So we saw that we we made two bets. Umthe first was dedicated silicon would bethe answer. The second was it couldn'tlook like a GPU.And our view as computer architects isif you want to be 20 times better thansomebody, right? Your architecture can'tlook like them,right? It it it can't. They have theyhave enjoyed and and eaten all thelowhanging fruit. So, if you build aGPU, the odds that you're better thanNvidia in our view are approximatelyzero. That led us to a fundamentallydifferent architecture. All right. Thehard part here, the hard part is movingdata from memory to compute.>> This is the fundamental problem in AI.And we solved it with a way that thatvery few others had

00:23:34even attempted,which was to build a very big chip andto put memory right next to compute. Andby building a big chip, a chip the sizeof a dinner plate, whereas most chipsare the size of a postage stamp, wecould use a different type of memory.And by using a different type of memory,a memory that was vastly faster, weopened up all sorts of opportunity.So when OpenAI uses us, we're 15 or 18times faster than a GPU.That means your answers are deliveredmore quickly. It means your engagementwith with the AI is more enjoyable. Itmeans you can use the AI to solve harderproblems and not wait.And the way to think about this is sortof to ask yourself the the counterfactual question. How big is the marketfor slow search today?Right. Right. Is zero. How big is themarket for dialup? It's zero. H how longdo you wait for a website to resolvebefore you click away? 3 seconds, 5seconds. You will not wait for AI. Wehave to deliver it to you in a in inreal time. And that's what we saw.That's what we built.>> So the panel's on going public. A lot ofLPs

00:24:50in the room. They need to getliquid. I'm curious about the journeyfor your investors. Yeah.>> Okay. So, well, you guys went publicwhat year?>> Uh 2021.>> 2021 by way of a spa.>> Correct.>> Okay. And your VCs were who?>> Um Drake Professor Jverson was one ofthe earliest Capricorn. Um Peter Te'sfounders fund. Then we got uh YuriMilner's DST.>> Okay. So your your investors come in,you go public at 2 billion via a spa.>> Now we're four years later. Really, itwasn't until year three or four that 90%of the value was created. Okay. So didthose early investors capture this 90%move? Did they stay in it?>> Most of them did. Yeah, most of themdid, which is really smart on theirpart. Obviously, I think they shouldhold on even more. Uh, if I didn't thinkthat, you should.>> I'm a little bit self-interested.>> What's interesting about this is>> No, but really they did. And and I mean,Google hasn't sold a share. They're ourlargest single investor. Um, uh,Capricorn didn't until very recently.So, basically, most of them stayedreally well in and they got all of thatupside.

00:26:05And good for them. And thereason I think this is so important>> is that there are a lot of LPs in thisroom>> who they're like when a company goespublic give us the shares.>> No no>> give us give us the shares. This is acounter example right? This happened tous in 10 years ago. We investedpreipo at a billion dollars. Wedistributed the shares I think at threeor four billion and then it went to 50billion over you know the course of thenext 24 months. And we had people whocalled us who said well why didn't youhold on to the shares? And we're likebecause you're pounding on us, right, todistribute to the shares. So you're anexample. Now>> in your case, Andrew,>> you have an innovation, right? You'rejust now uh public. So all of yourinvestors are still under lockup likelike Altimeter, but you guys haveinnovated with the banks on what I calla dribble lockup. So over 6 months theshares can be dribbled out according toa bunch of performance hurdles whichSpaceX is going to have a very similar>> when did we start this process of the ofthe dribble>> the dribble [laughter]>> concept you started it years ago but>> yes>> but u we're all of that with respect tothe lockup I think this is the the mostinnovative and I think

00:27:20SpaceX is goingto have a very similar innovation butAndrew for your investors if you weretalking to ILPs right in the room.Should Alimter be distributing theshares when they come out of lock? Howdo you think about you know your VCsholding on to the shares kind of postlock?>> Well, I I I think historically moremoney's made after IPO than before.>> Yeah.>> I I I think every single study showsthat uh there is more money to be madeboth in percentage and in in what wecare about which is absolute.>> Yes. and and and so uh the amount ofmoney that it's possible to put to workin most venture companies is verymodest. I mean there are two or three orfive outliers but for the most part youcan only put a relatively little bit ofmoney to work. Um by the time we getpublic there's a lot more money there ifthings are going well and theopportunity to make vastly more is afterIPO not before. Hey, if I could just addon that, one interesting question iswhat's going to happen with SpaceX onthis because you know a lot of the valueis is in the future, right? But I meanmost of the big tech companies wentpublic at a few billion,>> not a few trillion. Like there's a lotof zeros in

00:28:35between those, right? Andyou got all this upside afterwards. Now,for the equivalent liftoff, SpaceX wouldhave to be aiming at quadrillionvaluations. Now, I know Elon has thosesort of ambitions, but you really haveto believe in that uh to get, you know,>> this is this is kind of the point I'mgetting to, right? We have three megaIPOs, you know, we keep talking aboutthat are multi- trillion.>> Yeah.>> All of that value acrewed to privatemarket investors.>> Planet Labs is a great example ofventure capital in the public marketswhere the 10X has occurred in the publicmarkets. We're all advocates of thesecompanies coming public sooner. HadAndrew had his way, he would have beenpublic 18 months ago, probably at $10billion rather than $50 billion. Andthat 5x over the course of the last twoyears would have gone to public marketinvestors. So go ahead.>> Way better to be lucky than good.>> Yeah.>> So, so I think that I hear a lot ofpeople thinking that anthropic open AIand SpaceX are the new normal. Iactually think the public markets may beshifting back in this direction and alot of the companies in our portfoliosare now thinking about going public at abillion or three

00:29:50billion or 5 billion.>> We had this period of a decade whereAndrees was really pushing stay privateforever and I see the pendulum swingingback to companies are like man I want tobe like Planet Labs and get public,>> right? and and have to play in the bigleagues and do it in the public marketslike>> here's what I'll say maybe just like tothe two of you guys both of you guyshave had enormous pressure becausethere's visible competition that'salways sort of in your peripherybut I do think that getting publicsooner having the scrutiny of publicmarkets having the scrutiny of having todeliver sharpens the focus it steelsharpens steel iron sharpens iron>> and I think innovation tends to getbetter And so the idea that you alloweverybody to participate but you alsoput yourself in the spotlight to me iswhere great things happen.>> I agree.>> Um and so anyways I just wanted to sayto both of you um just as we wrap youguys are an incredible testament toentrepreneurship both of you. I meanwe've been talking literally since dayone me and Andrew because we went indifferent paths and then we kind ofrecon converged and then Will same withyou.>> I'm happy it worked out for you Jimoth.>> Well it's worked out for both of us soit's [laughter] fine. Um you guys areincredible testament to

00:31:05entrepreneurshipand uh I just want to say thank you foreverything you guys are doing and thenext few years are going to be reallyspicy.>> Yeah, if I could just spend a 30 secondson the next few years because I thinkit's going to be so exciting with um asI mentioned AI and space mergingtogether. We're going to see a take offour applications. I like to say like allthe cool stuff that we're doing on thewith LLMs now is really based on justthe text of the internet being absorbedinto these models which is incrediblypowerful already but they don't know about the real world. I call themblind to you know they don't know aboutthat farm field that flood that securitysituation around the corner. If you givethem real world data, then they cananswer real world problems. And that'sgoing to open up gazillions ofapplications for these uh AI models. Icall them instead of having largelanguage models, large earth models oruh instead of AI, planetary intelligencewhere you have planetary sensing systemsin space, planetary compute systems inspace. And we can disagree or agree onon exact time frame, but I think it'sgoing to happen. and and then that'sgoing to enable a huge economy. So, it'san exciting time in the next few years.>> Will Andrew, thank you guys very much.Well done. [applause]>> Thanks, guys. Okay.

00:32:20[music]Thanks, buddy. Appreciate you. Reallyappreciate it. Great seeing you,brother. Congrats.