Matt Levine: prediction markets are mostly sports betting, SpaceX IPO could suck oxygen from AI fundraising, private credit in AI isn't the 2008 scenario

um

we have Matt Lavine from Bloomberg and Money Stuff joining the show. He's in the ream waiting room. Let's bring him into the TV Ultra Dome.

Merry Christmas, Matt. How are you doing?

Merry Christmas. There you go.

Thank you so much for taking the time to come on the show on a Friday. Thanks for having me.

Uh great to great to meet you. Um

and thank you for creating wonderful content that at at times we have uh

remarkable. Yeah.

Many times included included in our show.

Triple glaze. Yeah. No, we love we love the work. It's been fantastic.

Uh I I I did want to start with a little bit of background. I I wanted to know about the day that you decided to become a journalist. What was the transition like? Uh what was like the first inciting element in that story? Did you know we were headed towards a hyper financialized world where we would be able to write about this very conversation?

Uh I don't know man I worked at I was working at Goldman when I decided to become a journalist. So I was I was already pretty hyper financialist but no I mean

that I was uh I uh the day I decided I I went to my boss and I was like I'm quitting to go to

Dealbreaker which is like a financial gossip blog and he was like that's fine. You can stay for another two weeks and finish up your work. And he was like, "Wait, did you say dealbook?" And I said, "No, deal breaker." And he's like, "Okay, now you're absolutely the book."

Yeah. We really That's remarkable. Um, well, I wanted to I wanted to take your temperature on a bunch of different topics. Um, the the the first might be uh prediction markets. I mean, Jordy was joking about it being able to gamble on this uh this very conversation. Yeah, we've had some uncomfortable moments throughout the year where people flood into our chat and they're, you know, betting on what Sam Alman is going to say or Mark Zuckerberg is going to say on one of our conversations and it becomes very distracting because everyone's just saying say this, say that. Obviously trying to uh

influence the

manipulate manipulate the conversation. Um but uh but yeah, it's been

I I get a lot of those people. I get a lot of um

yeah,

the overlap between prediction market people and my readers is pretty high. So there's a lot of people emailing me this.

Wait, wait. But so so so what I mean one of my main questions is like is like how do you actually get to some sort of ground truth for what's going on with prediction markets? Because there's there's a world where it was like, you know, this incredible crystal ball for understanding the election. Then it turned into, okay, this is going to give you information about all sorts of different things in the world. And then you peel back the onion and you think maybe there's just a lot of sports betting going on. like how do you actually think about the the shape of the business today and where it's going?

Um, you know, I think that like the stuff about like what people will say like you look at the volumes on these things, they're like, you know, thousands of dollars. Like it's not like there's a lot of stuff that is fun to write about because it's uh it's weird and silly and susceptible to manipulation, but it's just not that big, right? Yeah. Um, when people talk about these companies raising at huge valuations and becoming huge companies, like it's 80% sports gambling. That's like a number I made up. Like it's it's, you know, it may be less than that as a volume matter, but like it's it's like

the next leg up is sports gambling, right? I mean, that's what that's why they're businesses. I think if you talk to them, like their view is sports gambling is this huge opportunity for them to, you know, get people there, right? to get customers who really like gambling and to get uh professionals who can make money gambling against the the sort of like you know dumb money customers and then once you have that it's easy enough to expand into you know markets that make us more informed about the future of the world. Uh I don't know that that's true and like if they end up just being

um weirdly regulated casinos then like that's possibly still a really good business. Um

the question is how does the market actually evolve because you have every traditional sports book adding prediction markets and so

if you have new businesses that can be in 50 states on day one maybe that grows the market a little bit but overall you know what what is what is the sort of like total enterprise value that all the popular online trading andor betting platforms can really sit at is kind of unclear right maybe

yeah I mean Like clearly the big opportunity is like cannibalizing sports betting. The other thing is like it's not clear what they are because they're exchanges, right? So they're not making money by trading against uh customers, right? And so like one possible evolution of the market is you end up with people who are either literally sports books or who are like kind of like hedge funds that are replacing sports books being market makers on these platforms and sort of, you know, making their money by trading against retail blow. But like, you know, they're a separate business from the the actual, you know, Calian poly market platforms.

Yeah. Do you have an idea of how sophisticated like Wall Street is around sports betting? Because you have to imagine that like at some point Jane Street or Citadel said like, well, what if we just went and put someone in Las Vegas and started trading? Um, could we make an edge there permanently? Uh I I I it feels like there might be a wave coming where Wall Street starts having like serious desks dedicated to sports betting but or is that already happening or do you think that won't happen?

Yeah, it's already happening, right? I mean like Susahana is the big one who like you know famously have a sports desk and like other big kind of like high frequency trading market making firms are dipping their toes into it. Sure.

Um I think it's hard for them to

get into like traditional sports book,

you know, trading, but it's very if you're too if you're too regulated exchange,

if you're too good, you just get booted.

Traditionally, you get limited and it's like it's just like a different interface where and and a different regulatory interface, too, right? Like if you're trading on a commodities exchange, you're trading on a commodities exchange. They can do that.

Um uh so I think they're very much getting into it. And if you know legal sports books regulated as commodity exchanges are here to stay. And it's also just like uh whether or not it's a good business opportunity there's a huge personality overlap. Like you go to these high frequency trading firms like it's the same type of analysis. It's the same type of adrenaline. So like they they're very happy to get into sports books.

How are you thinking about insider trading on prediction markets? It feels like we've I mean this this week it seems like multiple people have been making the the steelman argument that it's good. Uh

yeah, the problem is is in some ways it feels like if you don't have insider trading then it's just a bunch of people guessing and if you do have insider trading then you potentially get greater accuracy of of uh and and like more value out of the markets and so it just feels like a very tough uh tough I mean it's very tough for the prediction market platforms themselves to actually talk about Mhm.

Right. Like my impression is that everyone who started a prediction market is a true believer in the idea that like the job is to make markets as informed as possible and like they all secretly love insider trading, but they're also trying to get approved by the CFTC and so they have to say we do not allow insider trading. I think the rules are weird, right? I mean the rules of insider trading in commodity markets are less strict than they are in securities markets because traditionally they're for like you know grain producers hedging their grain production. And so like you're supposed to insider trade.

Um and I think like maybe a little bit of that philosophy will be imported into the prediction markets, but I mean if you look around it like the Justice Department is going after you know baseball players for rigging gambling and baseball like no one there's no appetite for legal insider trading in most of the markets that uh that you know will will be big money in prediction markets. So I think that the regulatory situation is kind of going to shake out that insider trading is kind of prohibited but maybe like a little harder to monitor than it is in the stock market. But definitely philosophically people in this market don't want that. Is it do you think we just need a like a big like like like bombshell case or story to sort of like catalyze uh some action around that because it feels like sports is that

Yeah. Yeah. I mean I guess but sports books have existed for a long time. So I'm interested in understanding like what would actually change because because for what like I don't know 50 years it's been you know longer it's been for decades it's been illegal or you know against the rules for a baseball player to go and put a big you know bet against them and then throw the game and so that should just that I would imagine that that framework just carries over to prediction markets but uh maybe if it's lagging then it we won't actually get codified until there's a big headline in case.

Yeah, I think they be more codified in prediction markets, right? I think the prediction markets have rules saying you can't use material on public information whereas like the sports books the sports book have those rules too, right? Uh I think that I I think that it's going to be pretty codified like that, but you know, you're going to have a lot of like the sports stuff, the sports leagues have a real interest in policing and you know, the justice department has a real interest in, you know, that's a that's a sexy case to bring if you're like, you know, arresting baseball players. uh you know people betting on uh like you know Google search results maybe is a little bit less interesting and a little bit more lucrative.

Yeah, that makes sense.

And like also we like live in a world where like fraud enforcement is just kind of

less of a priority. So like people will definitely get away with stuff in the next few months.

Yeah.

Uh how did you process Trump saying he w quote would consider eliminating taxes on uh gambling winnings?

This just him joking around or something? Yeah, it was kind of offhand, but then the the the nature of the internet today is you have like a hundred accounts on X that are like pseudo news wire accounts and they'll take one line out of context and then blast it out and they'll get a bunch of engagement and then you realize, you know, you you you look to like an actual credible news source and you realize that, you know, again, fully fully taken out of context. But, uh, do you think we should be incentivizing uh gambling in the country?

Uh, so a couple things. one, no. Two, uh, you know, like like like being the friend to crypto has paid big political dividends to him. And like if you're sort of looking around for what other like

aggressive young male demographics you can can appeal to, you can might be like, I'll I'll I'll give the gamblers what they want. And then the other thing is there is this weird situation where the tax bill may be somewhat accidentally limited to deductions of gambling losses. So you can only deduct 90% of your gambling losses which if you are a professional gambler is like quite devastating. And one theory is that that has been very good for the prediction markets because there's an argument that if you you know are trading on prediction markets and you win someone lose someone you win 51% of your bets, you can deduct all of your losses. Whereas if you're doing that on a sports book, you can deduct 90% of your losses. And so it's the difference between being profitable and unprofitable.

Yeah. Uh, as far as I know, that was sort of an accidental consequence of all of this. But like, you could imagine a, you know, Trump like, you know, Trump administration liking the prediction markets and wanting to send business their way.

Yep. Uh, I want to talk about the potential SpaceX IPO. A lot of people have been processing it in different ways. Some people were surprised this would was happening just given that uh SpaceX has thrived in the private markets and has plenty of access to capital and and Elon has had a number of bad experiences in the public markets. I have a kind of working the thesis that this is could be a spite IPO. I think we're going to see a lot of IPOs next year and if you're raising a billion dollars, let's say if you're going to float at like a 50 bill, you know, if you're $50 billion company and you only need to raise a billion dollars, you're not really threatened by IPO. But if you're a foundation model lab such as OpenAI that wants or Anthropic that wants to raise tens of billions of dollars, it's very possible that a $ 1.5 trillion SpaceX IPO could kind of like suck the oxygen out of the room. uh is is uh you know I want you to kind of give us your take broadly on on it and and I'm curious if my uh tinfoil hat skitso theory has has any merit.

Yeah, I don't really know. Well, okay. So, first of all, like I you know, I read companies might go public next year all the time, so I don't I don't know. I mean, I assume it's true, but like I'm not totally sure. Um, two, like one thing I've read is that the use of proceeds for this hypothetical $30 billion IPO is to put data centers in orbit. And like you know this gets to your spite theory like in a world where there's you know10 trillion dollars of broadly speaking AI capex needs like there's going to be competition for that and there's going to be like like like SpaceX can kind of SpaceX has been able to raise as much money as it wants in the private market and like you know achieve a you know $800 billion valuation and like send rockets into space and be like a fairly capital intensive company while staying private. But like if the next leg is, you know, they need tens of billions of dollars of AIish spending, then uh you might reach the limit in the private markets. And yeah, you might reach the limit in the IPO market, too, right? Like you might like there might just not be enough money to go around for all of these AI companies. And then I don't know why SpaceX is in the AI company category other than this.

Well, so so here's here's the other Elon complex, right?

Yeah. But the other theory is like there's a world where XAI could get rolled into SpaceX and taken public around the space data center narrative. And I think that

there's a lot of people that would be excited about that narrative even if other people in the space economy are saying, "Hey, space data centers will happen." But it's probably more like a a 10-year timeline versus a 3 to four year timeline.

Yeah. Also, like I mean I've read that they're like looking at a $ 1.5 trillion valuation on like $22 billion of revenue. Like it's all

pretty it's all on a pretty deferred timeline, right?

It's an orbit. It's an orbit.

Yeah. You're not paying for like that year's earnings, right? So, you need you need some sort of 10ear story to tell.

Yeah. But I mean to be fair, it would be easier for I mean at least I believe it'd be easier for SpaceX to acquire XAI than Tesla to acquire XAI because of acquisition laws around a public company buying something big versus a private company buying something big. Does that sound roughly correct? Like would you handicap?

Oh yeah, it sounds roughly correct. Although I mean Tesla has form of buying,

you know, other Elon companies and

yeah,

you know,

has moved to Texas in order to make it easier, right? Like I don't think Tesla would have that much trouble buying, but they would get sued, you know, and if you do it in SpaceX, it's like

Well, I think I think part of the challenge is that Tesla shareholders might say like, "Hey, we already have a really good AI team. We've been working on self-driving. Like, we don't need XAI. We don't necessarily want to dilute ourselves, you know, whatever XAI's current valuation is or whatever price it would come in at. So, I I just think

I guess I just feel like the sort of standard Tesla shareholder approach is like we're going to trust Elon's intuitions on these things and I mean they bought like you know

Solar City

solar company, right?

Yeah. But there w there still was push back and that was like a much I'm assuming that be 1% of the that'd be one like Solar City was like 1 point something billion, wasn't it? It wasn't like a smaller.

It wasn't like a 100 billion. Like I'm assuming XAI has a lot of debt. They've raised a lot of equity. It would be

somewhat.

It has the Twitter debt as well.

Oh, yeah.

It has a lot of debt rolled in. Uh well, speaking of debt, I mean, Netflix had to issue some debt for this Warner Brothers deal. I'm interested to know um your your thinking around how the Warner Brothers uh process will go. Are we going to wind up in a quiet period for the rest of the the year as we go into the holidays or do you think it'll be uh you know just as as uh just as many headlines as there have been? Where do you think it actually lands? What's the nature of the story?

I don't know. It's been surprisingly quiet to me. I think it's weird that um that Paramount first of all launched a tender at the price that was rejected, right? rather than saying we're going to raise a few bucks and go to the shareholders, they launched at the price that the board said no to

and that they've said publicly that it's not their best and final offer. Um

crazy is there any is there any logic where that makes sense?

Okay, so I think that where that makes sense is that um is that they're it's not a real tender offer, right? Like they can't close a tender offer for

reasons including antitrust, but they clearly want to do a friendly deal with the board, right? Okay. And so I think that saying very clearly, we could give you an extra two bucks if you sign a friendly deal with us is a is a good way to get, you know, to the board and to try to strike a deal with the board. Um, so I think that's the logic there, but uh I I don't I don't know what the board is doing, right? I mean, I think the board likes the Netflix deal. I think that I also think that like um Paramount has two things going for it. One is that they think they've offered more money or certainly they can offer more money and then the other is they really think the Netflix deal is dead on sort of antitrusty reasons.

Yeah.

And I don't know how that plays out, right? Because that's a that's a potentially year-long process and presumably nobody wants to wait that long. I think like there's some ability to wait a few weeks and see if like Trump says nastier things about Netflix to try to, you know, pressure the board to come to a deal.

I I don't know. I'm I'm surprised that there's been so little news, right? I mean, like since Monday, like no one like

I would have expected someone to raise their offer by now.

It's a good point.

Yeah. Uh what was your Have you formulated a take on the uh OpenAI Netflix deal? We kind of landed on mean Disney deal

or sorry yeah open AI Disney deal. Uh we sort of landed on it being like

sort of an interesting way to create less commoditization in uh you know the AI application layer the consumer chat experience they all are starting to feel very similar. I don't know if you've felt this way but feels like all the models are you know sort of at the same level. The benchmarks are pretty similar. Well, if you can get Spider-Man in one and not in the other, maybe that's a differentiator. Have you have you processed that deal similarly or do you have a different frame of mind around it?

I haven't thought that much about it, but that seems right to me. And it is like, you know, it's like, you know, you think about like the Netflix uh Warner situation, right? Like you have like

you have like content distribution services and then they have like content libraries, right? And there's some combination of acquiring the content and making you know commercial deals with the content providers right like like here one possibility is that the AI chats are ultimately the next you know dumb entertainment channel right and so if you have the good properties in your dumb entertainment channel then like when people are like oh I want an AI to create a movie about me I'll go to the one that has like the Disney characters like that I don't know it kind of makes sense right and it's like a um like the thing that is striking to me is that uh the AI companies like talk about like this sort of fundamental reshaping of society and you know discovering super intelligence and then like they're a little bit consumer entertainment products but like if you're a consumer entertainment product like you want the Disney characters

also adult adult entertainment.

Yes.

Yes. Right. Right. Disney character porn. Well,

it really has.

You're saying it is it is funny that adult adult mode and Disney are will be rolling out in the same effectively 30-day period. So,

I don't I don't know enough about the deal to know if they like have um I hope they have

I hope they have a carve out.

They do have a carve out. They do. This was confirmed. They did say that there are lots of restrictions on what you can and can't do with a Disney character in OpenAI properties. I'm sure that was that was key. Uh

it's wild. Um

we're gonna we're gonna see the jailbreaks. That's gonna be good.

There will be some jailbreaks. I think there won't be that many jailbreaks. I think in the last two years they've gotten pretty good at locking it down. So that was that was kind of my my prediction. Uh I'm sure but I'm but there will always be chaotic stuff. It's it's it's inevitable with AI, the internet. Um what uh uh how should tech people be thinking about the growth of private credit? I feel like there was a uh an era where folks in tech were you know understanding that the foundation model companies anthropic open AI were operating at a truly unprecedented scale. The capital war was bigger than Uber Lyft. It was bigger than anything uh folks in tech had seen. It was certainly more capital consumption consumptive than Google was when before it IPOed. And so all of a sudden, by a lot, so all of a sudden you have tech people that are grappling with um the prospect of a10 billion dollar debt financing for something that is, you know, feels like a startup. It feels like a startup project. Uh and so it was easy to point the to sound the alarm bells like debt is coming. Debt is scary. Debt blew up the economy in 2008. It's going to happen again. Uh but you talk to a lot of people in private credit and they say no there's plenty of areas where you know people pay their pay their debt obligations on time. How how have you processed like the is Nvidia holding up the global economy? Is open AI going to blow up the entire global economy? Those types of like Cassandraesque narratives that have kind of spiraled and maybe are coming back down to earth now.

Uh yeah, I don't know. Yeah, I mean I it's fascinating that you know like insurance companies investing in investment grade debt are like ultimately underwriting like a stream of payments from startups, right? I mean like that's kind of what is happening here.

Deeply unpr deeply unprofitable deeply unprofitable startups.

Yeah. But like also a lot of it is like people underwriting a stream of payments from Facebook which like you know um

uh

is very profitable.

Yeah. Yeah, like the blue al data center for for meta super intelligence one gigawatt the Hyperion thing like it's like Meta's gonna pay that bill at least I think so

right and if you had made like an equity bet on like if if there had been a way to make an equity bet on like Meta's pivot to the metaverse like you would have lost all your money but like like presumably they paid their supplier bills for that right like

oh totally

like there's a there's like a difference there if you're like underwriting these like hyperscalers like mostly um

you know these are very investment very cash flowy companies that you know could could turn off the capex if they needed to. But right, I mean you're underwriting a lot of you know fixed payments on capex and uh it's it's um uh it seems a little aggressive for a credit investor, but um they're all like one thing about private credit is like uh it's not going to blow up the economy. like it's you know it's like insurance companies making these bets so it's a little bit longer term capital.

Yeah.

But I don't know.

Is there anything in the economy right now that gives you the same sort of jitters that you had during your famous interview with Sam Bankman Freed where it felt like at the time you were clocking like this just doesn't make sense. The math doesn't math. Like it feels like we're frothy. where everyone kind of goes around saying like, "Yeah, we're in a bubble, but it at the same time it doesn't it feels like it's a bubble that's built on some real stuff." Like even even the prediction market stuff, it's like I can at least trace the cash flows. I might think that sports betting isn't the greatest thing, but like at least I know that the businesses are going to make money,

right? Like you know, AI sort of replaced crypto as the sort of darling of a lot of investors, but it's just you can you can the the consumer product is very tangible, right? like the crypto stuff was all like, "Oh, I'll make tokens that other people can trade for more tokens and and and like it didn't

Wait, that actually sounds like AI too because I make sure

I make enthropic tokens. I sell them to you. You make cursor tokens. the circular like the circularity is is you know if you want to be troubled by the circularity it's like a little weird but it's um

the difference between that and like crypto where no one like like the thing about Sam Bankman Freed that was obvious at the time was that he didn't care about crypto he wasn't like oh yeah this is a world changing product it was like oh yeah this is a casino where people will give me money um

and like AI is there's a lot of people who are true believers in it and again they've like rolled out consumer products that have Disney is going to license its characters, right? I mean, like there's a real um uh sort of there are many real obvious use cases and so like there's a lot of value being created and then the question is like is it commensurate with the you know $10 trillion of debt that's being incurred and that's uh it's not obvious but it's like you know there's there's something there. There's not

how much how much would I have to how much would I have to pay you for you to agree to not use any AI for the next year? Um, I'm kind of a let like I mean I use AI but uh uh

I'm assuming it's like for re like research here and there of like better kind of search.

No, give us the prompt for your column.

Give us the prompt.

What's the prompt?

People pretty regularly email me. They're like, I asked Chat GPT to write about this in the style of Matt Levian and it did pretty good. And they sent it to me. I'm like, okay, I'm safe for like another three months.

Yep.

It's like never

never close. No, it just never just doesn't get the flavor or something. I don't know. May maybe

the answer to your question is less on you than like do it for

Yeah. Yeah.

How have you been processing Oracle over the last six months or so? They they're getting zero credit for their backlog, their five half a trillion dollar backlog. Uh I think they're are they trading at less than less than the backlog currently? uh they you know people have brought up uh the dotcom bubble and obviously are drawing comparisons between the two moments but what's your view

uh about Oracle being like not the darling of the IO bubble? I I don't I don't really know.

Uh uh they're just not I don't I don't I don't know enough about like Oracle's backlog, but I I think they're not just, you know, they're not the name that that people go to as the as the AI darling. Um

yeah,

I don't know. the.com bubble is like, you know, it seems like a reasonable comp in that. Uh I don't it just feels like obviously even if there's a huge shakeout there will be huge value you know there'll be an Amazon of this bubble right like there'll be big important survivors of it which you know again like I can't say wasn't true of crypto but like there's nothing inevitable about that in crypto right like that was you know like like there's there's there's the possibility of nothing happening there whereas is here. It's like like the changes to the economy and to society are kind of already tangible.

The thing that's somewhat of a narrative violation is the like the median crypto fund has actually done pretty well over the last few years. And it's possible that the median

AI focused boom. It's like you got to be in Google or Amazon or you're probably in a lot of trouble. And and uh and with crypto there were a lot because it's so liquid. There were so many funds that were like trading in and out. Even if something went way down, it's like, well, they already distributed. They got out early. Like, you know, there's tons of ways to make money.

Yeah.

I don't know.

Yeah.

One way.

Yeah.

One kind of random question on your workflow. I'm assuming people email you kind of like tips tips all the time that are really trying to like feed you information on on their enemies, things like that. M

how do you kind of uh

I'm not an investigative reporter and so like I don't I'm not that useful for that but yeah people are like

you should write about that and

but how do you yeah what's your call this process

uh what's your what's your process for trying to figure out what that is is like actually real because I mean we we

we get that where people are trying to like convince us of certain narratives and and we're trying to process you know maybe off air trying to process how real something actually is.

Um, you know, again, I don't like do a lot of investigative stuff. So, if people are like, "Oh, look into this secret thing." I'm like, "No."

Um, so like they, you know, like if I'm if they're sending me information about their enemies, it's like, "Here's their 8K filings and I can read them myself." Um, and I do try not to I used to like find it flattering when like,

you know, big finance people were like, "Let me tell you about this." And then I was like,

"They're just shopping me their version." And so I try not to

Yeah,

sure.

I try not to either have those conversations or if like I have conversations with the hedge fund manager about how their position is so good, I just don't write about it because

I don't I I it's like too easy to be to be um flattered and persuaded into thinking that what they say is right. And so like I just try not to expose myself very much.

It's good.

Yeah. Like I just try to you know like like people like oh this is an interesting like okay send me the public documents and I'll read them and I'll make my own thoughts about it. You know there's a balance.

Somewhat related question.

You have to get some information from people but like I try not to be

hearing their opinions.

Somewhat related question. What is your theory on why Bloomberg is able to retain top journalistic talent? They've seen very little churn over the last couple years relative to other legacy outlets. What's going on? What's the secret sauce? Um, I mean there's a lot of things like we have a great audience. Like the audience is so good. Like it's so, you know, this like the feedback that I get from readers is like they're so smart. Like it's it's so concentrated in the financial industry. So like the audience is so good.

Um, it's the best business model in media. Like we're a good, stable, profitable company in a world where that is not true of a lot of legacy media. Um and uh I mean and you know and there's a flywheel where like your colleagues are great. Like I really like like I wouldn't leave Bloomberg just because my friends are all here.

Yeah.

And then the other thing is like it is a huge organization that nonetheless gives people a lot of freedom to do weird stuff. like what I do uh like I've been doing this for like 10 years and you know if I tried to do it at like another media organization they would have various statistic objections and I'd say are you sure you can do it this way? Like I feel like I proved out the model and so Bloomberg just lets me do it. But like Bloomberg does that a lot, right? I mean like OddLots just had its 10th anniversary party this weekend. Like OddLots is a weird

overnight success.

Yeah. It's like a program that just, you know, like a lot of people in media would have been like, "Well, that won't work." And Bloomberg just let them do it, right? So, there's a lot of stuff like that where it's a huge organization that also is is like quite

experimental,

agile, like Yeah. And like gives people a lot of freedom.

That's really cool. Yeah, it's a great that's a great take. Well, thank you so much for taking the time to this.

Really enjoyed uh

Yeah, this is very educational

letting uh letting us bounce bounce our theories off you.

Yeah,

this was cool. Thanks.

Well, have a great rest of the year. Have a great rest of the day and we will talk to you soon. Thank you so much.

Cheers, Matt.

Goodbye.

Finn.ai. It's the number one AI agent for customer service. Automate the most complex customer queries on every channel with Finn.ai. Our next guest is