Meta Debuts Muse Spark 1.1, Blue Origin to Raise $10B, Cursor Develops AI Agent

0:13Welcome everyone to the Informations TI TV. My name is Akash Pasricha. It is Friday, June 10th. Before we get to the show today, South Korean chipmaker SK Hynix is set to begin trading on the NASDAQ today in its U.S. debut. It is a new major proxy for how investors are thinking about the memory chip crunch. We'll have more coverage of that on our website, so check it out. Today on the show, we're going to unpack Meta's latest AI model update, Muse Spark 1.1. Early reviews seem strong, but will it be strong enough? We'll get to the bottom of that. We'll also break down the pricing angle of all the new models that were released this week in our weekly edition of the Editor's Cut.

0:56We're also going to understand more about Blue Origin's new funding round and what that means for the space race. And we'll close out the show with our exclusive reporting on how Kersher is developing its own AI agent to compete directly with Anthropik's Claude co-work. It's going to be a great show, so let's get right on into it. Meta announced the release of MewSpark 1.1, its upgraded AI model made for complex coding and autonomous tasks. The company has been seen as a bit of a laggard in the model race, but Mark Zuckerberg has invested heavily into fixing that. Joining me now to discuss all this is Andrew Boone, a managing director at Citizens.

1:38Andrew, welcome back to the show. It's great to have you here. Great. Thanks for having me. All right. So let's talk about MuseSpark. What were your reactions to it? I think it's a step-function change in terms of how you think about meta over the next kind of five years. Last night, I think there was reporting out that Meta wants to treat its API of its frontier models as a standalone business that can support itself. If you think about that and, you know, kind of what are the building blocks for becoming a frontier lab, right? It's really energy, data centers, tokens, and then kind of what does that create in terms of intelligence?

2:17I think Meta has the building blocks to be able to go there. It's just the question of what level of intelligence are they able to create and what does that look like versus an open AI or anthropic. To that level, Meta's done a really good job of trying to catch up here. They've accelerated what is the release schedule of their kind of frontier models. And I think last night was a pretty good update. And so overall, I think we're pretty positive about kind of what Spark looks like and really the creation of what could be a new potential business line for Meta overall. And this is kind of interesting because Meta has tried for many years now, many different ways to develop an enterprise business.

2:56And it hasn't quite caught on, I think, the way many would have hoped. I'm remembering, you know, Clara, she was an executive that came in and tried to build an enterprise business that way. So this sounds like there's more hope than ever for selling to businesses something other than advertising. Yeah, I mean, let's go backwards. So they acquired a CRM that was customer. They've done things like AI business agents. I think all of the tools that they're offering to SMBs, it makes sense, but it's all in kind of, it's to help the advertising business. And so if I think about what this is, I mean, this is a developer tool.

3:38Like this is a whole different segment that Meta is going after where enterprise used to be hand in hand with advertising. Don't get me wrong. I do think that Spark will get distilled down into advertising-specific models, and the benefit of advancements in AI overall will help the advertising business. But this is an entirely different product set in terms of what's meta-building, and that in and of itself is just incredibly interesting. So I want to ask you about if you think this is a smart strategy, because AI and meta, if you go back maybe a year ago, I mean, Mark Zuckerberg, he talked about personal super intelligence as his vision for how AI makes its way into daily life.

4:21And that was more of a consumer vision, I think, that he was painting. you know whether or not we've gotten to personal super intelligence i mean the the glasses are getting better there's no doubt about it and you know the um the display ray bands they things will get better over time but do you think that it's smart to diversify here with enterprise while the personal super intelligence thing is still being developed i think it diversifies the bet. And so there's a bunch of strategic kind of off-ramps if you think about what the sale of an API for Meta's models means. And so one is, does this mean that Meta can then invest more in terms of data centers, energy agreements, everything else, because they have another outlet in terms of what they could utilize the compute for?

5:12I think that's one. Two of which is, can you amortize the cost of training models over more products? I think that's also true, right? We talked about advertising earlier i definitely think there's going to be some form of meta models that get distributed via the glasses right and like we can talk about that but that could be the next compute platform right if you think about what open ai released this week in terms of voice voice continues to get better and is that now more of a capable ux in terms of consumers interacting with models in a new way with agentic as the back end that really implements what is the action like All of that's really interesting.

5:49But the idea here, big picture, is that you now have a wider array of products to displace that cost, right? And so if training models is now billions of dollars, why wouldn't you sell one more product that you can allocate expenses to? Have you modeled out how big this API business could be for the company? I truly don't know, right? Because it's going to be a question of kind of like what developer tools look like. And right now we're in private preview, right? And so like, it's very early. But is that an exercise that you might ask your team to go back and do? I mean, is this going to be that big a business that you're going to need to model out sort of that impact on the top line for the company?

6:37Yeah, I mean, we took a cut out of last night, right? And it was as simplistic as just saying if Anthropoc does, you know, 30 billion and they're on a pace of what's been basically 10x growth, like what does this look like? Can meta be 5 % of anthropic revenue? Like that 5%, that's a billion-dollar-plus business right there. I don't think that that is that much of a stretch, and especially with the aggressive pricing that meta is throwing out there. That being said, you really want to be at the frontier of intelligence to win what is the share of API calls that are being placed by developers. And so the trajectory of Meta's intelligence is improving, and this was a really good reinforced view that Superintelligence Labs is in the right direction.

7:25But it's really hard to tell you what share going to look like in 2028 as we think about where the models end up in two years. The market's just moving way too fast. There's another side to this story, which is that as the models have been released, we've done some reporting on the extent to which employees have their screens being recorded and stuff like that, the extent to which that's being used to train the models. I think it's sort of understood that that was at least a decent-sized chunk of the training that happened. There's also the narrative here that you have to opt out I think on Instagram of having, if you have a public profile, you have to opt out of having your content being used to create AI generated content.

8:20And, you know, I guess the broader question I have on all this is, this is the approach that Meta has taken. They have taken a very aggressive approach to making their tools better. Do you see this coming back to bite them at all in any way? Is this just the game that we're in? What do you think? I fundamentally don't think that consumers care about data privacy. I just don't think that you've seen any sort of inclination across consumer behavior that tells me that they're going to turn off XYZ service because of data. There's a small handful of people that are very vocal about that, but at large, I just don't think that's a thing.

9:02Let's step back and talk about what is the asset of what is a media company like meta or honestly google with youtube and think about that within the context of open eye open ai anthropic and others right there's a really interesting data set that ends up being unique to both of those players because of the user base they have and how to actually think about training models with that information and so you know again let's bring this out more broad. It really highlights what is the uniqueness of kind of Reddit's data, who should be up for renewal in the back half of this year, as we extend out like, what is the thought of proprietary data and how important is it?

9:42I think it's pretty important as all of these labs try to suck up everything they possibly can to improve their models. Let's go to Google, though, for a second. I mean, you brought them up. They seem to have fallen behind a little bit in the model race uh and you know they've got a lot of stuff going for them with the application play and and uh i mean the coding model i guess people were a little bit underwhelmed i guess by by where they're sitting the coding race but my prediction is that they're going to come back in a big way and look we're talking like months of step change here right what i'm saying today will change a month from now and so it's not like they're terribly behind And I think they're going to come back in a big way in the model wars here.

10:26That's my prediction. What do you think? The broad expectation, right, of kind of like the market, everyone that we're talking to, it seems like everyone else that's out there, is that the next model gets released a week from today, right? So it feels like it's coming next week. And the question is going to be, what is the step function change for Google? Again, trading is somewhat silly to look at on a near-term basis, but market has rewarded meta by about 10 percentage points of stock price appreciation just from what is an updated model that appears to be a significant improvement. I think Google has been punished similarly by market participants because there's been a view that Gemini is slowing down.

11:10You've definitely seen some people leave. I think that's your guys' reporting in terms of people stepping away from DeepMind. Yeah. But overall, right, you can't count Google out. They have so much compute. They have the history of... I mean, they have the data, too. Like you said, I mean, YouTube... Exactly. They have it all there. And I've actually... I don't know the answer to this. I've thought a little bit about who has the upper hand in data, whether it's meta or YouTube, because they have so much at their fingertips. my hunch says it's still Google, but I mean, what do you think? I would just assume that it's Google, right?

11:50I mean, YouTube is just this tremendous database. I'm trying to think of what the old stat was that it's like, you know, 500 million hours of video is uploaded every minute or something. It's something insane. I was trying to think about how many people are on Instagram, Facebook. I guess that matters, you know, private accounts and stuff like that so maybe it's not it's less about that as much as it is about the actual content itself right and i i would argue that i think the content within youtube where it's like how do i fix a sync is very valuable right and so like you have something that is a dedicated video that is pointed at a subject i think that's a little bit different than the social media content that we very much enjoy and engage with than ends up being on Instagram.

12:39Right. Last question for you. So with all this in mind, are you expecting Meta to continue increasing CapEx at the rate that they've been investing it? We took up our CapEx numbers last night. Okay. And so if anything... How much? We're now at$200 billion for 2027. And so you talk to kind of buy-side investors, you talk to kind of where everyone else is. I think the upper range is now probably 225, maybe 250 in terms of what 2027 could look like. This is meta is also investing in leases, right? And so that's off balance sheet, non CapEx type investment spend. Point being is it's massive. You know, you got the I think it was the seven gigawatts that are expected now for next year.

13:26Right. If you just ballpark that and you say, hey, look, it's 50 billion a gigawatt that you end up putting in. And again, that's not a one-year period. But we're talking about a lot of money that, again, is being turned into capacity that should be able to help fuel not only advertising business, but also API sales in the future. So I don't know if 27 is the peak year, right? That's a lot of money. And we're now at a negative free cash flow kind of number for 27. I think it's down$40 billion, give or take. Like, how much can that sustain if they move to 28? I don't know. Is the rate of that, is that faster or slower compared to the other hyperscalers?

14:10You know, Meta is not a hyperscaler, but we consider on that. I mean, Amazon is 200 billion for 2026. Google is like 185 for 26. And so we're at Meta now being at 2026 levels in 2027. I can't see any way that you don't have Amazon and Google both tick up their CapEx expectations for next year as well. The AI CapEx cycle is very much still alive. And if anything, you see Grok also put out a model this week that's good and cheap. I think you have five hyperscalers that are all going full bore in terms of the AI frontier model race. And that means spend is going higher. Great. Well, I want to thank you for coming on, Andrew.

14:57That is Andrew Boone, Managing Director at Citizens here on TI TV. Meta wasn't the only company to release a new model this week. OpenAI and XAI also had big releases. Pricing is a big differentiating factor a lot of people have been paying attention to with these models. For more on that, I want to bring on our San Francisco Bureau Chief, Jason Dean, and Features Editor, Nick Wingfield, for this week's edition of The Editor's Cut. Welcome to you both. It's great to have you here. Good to be here. Okay, Jason, I want to start with you. So what did you make of all the model releases this week? What were you paying attention to?

15:33Well, a bunch of stuff, but I would say one of the more interesting facts is the price competition is clearly intensifying. So the GBT 5.6 model is about half on a per token basis, depending on if you're talking about input or output. And there's some independent analysis that puts the cost sort of on a performance basis at about a third. That's both relative to Fable 5 from Anthropic. And then you have Meta coming in for the first time offering an API for developers for its models. You know, it was doing all open source. Now it's trying to get in this game of charging people to develop using its models.

16:17And we haven't seen the same levels of analysis of the model yet, but Zuckerberg came out and said they tend to be very aggressive on price and compete there. So I think this is not a huge problem for Anthropic yet because they still have so much demand and there's a lot of demand at the cutting edge, people wanting to use the bleeding edge models and coding in other areas. But medium to long term, I think that the price wars are only going to get more intense. All right. Nick, what stood out to you this week? A couple things. The first is I really feel like we're in that phase before Uber went public when it felt like the company was subsidizing ride shares in order to win business from Big Taxi.

17:11Do you remember those days when Ubers were inexpensive? expensive that kind of feels like the phase we're in right now with uh potentially with uh yeah with this api pricing and now there's absolutely no discount at all they get an uber it's like the most expensive thing in the world yeah totally the other thing is um uh this is kind of an information war that uh that met as in uh nothing speaks to the importance of influencing kind of lead opinion on this than Zuckerberg breaking his Twitter fast over the past few years and actually posting and promoting this model on Twitter, or X, we're calling it now.

17:54It's really clear that they need to send a message to influencers, let's call them shareholders, as they embark on this journey of trying to win business from OpenAI and Anthropic. They need to shape sort of public opinion, elite opinion, because it's going to be a very, very expensive battle. Right. Well, you know, speaking of X, Nick, you know, I was, Elon's, Elon was in a pretty supportive mood this week on X, it seemed. I mean, he was praising Anthropic. And then And when Zuckerberg tweeted, you know, I think he was, I mean, he basically, why wouldn't you promote it? You know, saying Mark Zuckerberg is on the platform.

18:41But, you know, he was very much taking a more supportive tone this week in terms of all of his buddies, which is sometimes not the tone that he takes. Right. Well, Amprofic is a customer now, right? Or a client of SpaceX. So you can kind of explain that from that position. But yeah, maybe he's feeling in a more magnanimous mood now for some reason. Yeah. Jason, I want to come back to you here. So we were talking with Andrew Boone from Citizens before you guys about the enterprise strategy that Meta is going for here. Meta has struggled in the past to really turbocharge its enterprise business outside of selling ads, really.

19:29The API business is an interesting opportunity here. Do you think that this becomes a significant revenue business for the company? Is this finally the big hit that they were waiting for? I think there's plenty of reason for skepticism there. I mean, it just doesn't seem to be in the DNA. And, you know, that doesn't mean people can't reinvent themselves, I suppose. but they've been trying in enterprise and in other areas with virtual reality. And these have proved costly distractions from the core advertising business, which is a great business and is benefiting a lot from AI. I just, you know, to try to compete with Anthropic and OpenAI in enterprise AI products and models and stay at the bleeding edge there doesn't make a whole lot of sense to me.

20:26I think the idea of becoming a cloud provider, which is also part of the sort of rumored and reported enterprise ambitions there, also seems like a stretch. That said, Elon Musk has sort of demonstrated that you can do big deals for cloud and rent out your GPUs to Anthropic and others who are desperate for capacity right now without building out a whole sort of enterprise infrastructure inside your company. So perhaps there's an opportunity there for sort of short-term billions of dollars in revenue for excess capacity that they haven't grown into yet because Lord knows they're spending lots and lots of money to build out their compute and maybe building that faster than they can use it.

21:18But the idea of like Meta turning into a serious enterprise player seamless, like, a stretch to me. Okay. Nick, do you have an alternate opinion on that? I'm actually a little more bullish on this. I think that we are just one... Look, at the end of the day, Meta has all this talent, right? We are one model away from everyone wanting to use Meta's model, certainly in coding. I think they could do it. Nick, what do you think? Okay, well, there's a real one-trick pony issue when it comes to any company, any big tech company. And Amazon proved many years ago that they could branch out from the consumer business, get into enterprise with AWS.

22:01We all thought that was like a one-off situation. Then along comes Google after years and years of trying, and they managed to pull it off too. So I don't think there's anything inherently impossible about this for Meta, but it's not easy. But if you're Mark Zuckerberg, you probably look at what Google, which is your closest peer in the advertising business has done in terms of Google Cloud and say, I can do that too. Right. Jason, coming back to you, I mean, this was, well, okay, what were you going to say? Go there. I think it's a fair point, but Nick, you also just mentioned like the only two companies that have pulled that off.

22:44Microsoft's been trying to be both a consumer and enterprise company for a long, long time now, and is basically failing. The only real success in the consumer side historically was Xbox, and that is now a total mess that they're trying to figure out what to do with. I think also Google, just as a company, has a longer history of being fairly effective at trying a lot of different things. I mean, obviously their core business, the thing that made them money, has been advertising for sure. But if you look at all of the other bets, many of which didn't pan out, but some of which did. Waymo, for example, panned out at least in terms of technological and market distribution success.

23:27It isn't yet a moneymaker, obviously. So I think, yes, it is possible. I don't know that we've seen anything in the rather long history now of Meta to suggest that it has that ability and also to what degree is all this a distraction and i'm not talking about building models i think there is a utility to building models but to try to be a player and build out all the things that go with an enterprise business in terms of sales and customer service and go to market etc um as opposed to focusing on the benefits you can reap in the core advertising business which is only going you know the opportunity there is only growing um you know There may be an opportunity cost in the resources they're putting into those other things, given their lack of a track record in proving success.

24:16Nick, I want to turn to another story this week. Blue Origin is raising money for the first time, and it's not a small funding round. It's$10 billion at a$130 billion valuation. I believe that's before the fundraise. What do we know about how this funding round is coming together? Well, what I reported is$4 billion of that is coming from KOTU. So that's their first outside investment. They have, of course, been raising money throughout their history, their 25-year history from their family. 25? I didn't even realize it was that old. Man, 25 years Blue Orange has been kicking it? Almost 26. Quarter of a century.

25:01What was the... Was it always space rockets even 25 years ago? Yeah, yeah. I mean, they started out... I mean, people forget this, but they actually had a reusable rocket before SpaceX did. They actually demonstrated it before. These two companies have been kicking it around for a quarter of a century. So, Jeff Bezos, that whole time, was funding it. Now, in the early days, it was not as expensive to fund as it later became. But we don't really know how much Bezos has spent over the history of it, but certainly billions, tens of billions, I don't know, possibly. But at some point, Blue Origin has to become a real company with revenue and ultimately with profits.

25:50And I think we're getting closer to that now. And that's why they're raising money. I mean, there's some other factors here as well. The SpaceX IPO clearly has shown and generated a lot of excitement among investors for space. So that probably is a tailwind for Blue Origin in terms of raising money. And they also have contracts. And they've got a data center in space business called TerraWave, I think, that they're trying to push along. So I think that's why. And, you know, Bezos probably wants to stop subsidizing this to the same. He's putting$2 billion in this, to be clear. So he's not backing away from funding it.

26:35So Jason, I mean, what do you think? Do you think they try to go public in the next couple of years? Is that the endgame here? What do you think? I can certainly see that possibility. I mean, I think for now this sort of funding round makes a lot of sense for the reasons that Nick said. They have a long way to go, though, to catch up to where SpaceX is. I mean, it's striking. Nick was recounting, you know, they were first to build the reusable rocket. They've been around for a quarter century. And look how far behind SpaceX they are. You know, they don't really have an operating business now. They have revenue, but it's all built on sort of plans, contracts for things that they expect or hope to do.

27:18So, you know, a lot of people thought that the SpaceX valuations at the IPO were a little extreme. I think in some ways they look reasonable compared to the valuation of$130 billion for a business that isn't really operationally generating revenue yet. yet. So maybe you can certainly imagine they would want to IPO for capital access and other reasons, but it seems like they've got quite a ways to go to catch up to Elon. And Nick, you mentioned those contracts. Can you just remind us, you know, how big are those contracts? Who are they with? What are the main business lines that are actually generating revenue?

28:03You mentioned Orbital Data Centers too. I'm not sure where that's at. Yeah, well, they've announced a bunch. They've deployed a satellite from a client, and I unfortunately can't remember who it was. There was a bit of a mishap when they deployed it. But they do have contracts with NASA to deliver both astronauts and lunar rovers to the surface of the moon as part of the Artemis program in the coming years. And then I think they've got a variety of deals with telecoms and things like that to deliver satellites. So that business is finally, you know, kind of building up. And then if they can actually launch their own, you know, satellite internet service and other, or their own data center and space business, then that would be another source of revenue as well.

29:00Great. Well, I want to thank you both for coming on. And that is Jason Dean, our San Francisco Bureau Chief, and Nick Wingfield, our Features Editor, here at The Information. Cursor is building its own AI agent, according to exclusive reporting from The Information. I want to bring on our Elon Musk reporter, Grace Kay, who reported that story. Grace, welcome back to the show. It's great to have you here. Great to be here. So what do we know about this new agent that Cursor is working on? Yeah, so it's supposed to be a personal agent, kind of similar to Claude's co-work. So it would have access and send emails, send texts.

29:36It could organize spreadsheets in addition to performing, you know, some engineering tasks. So it sounds like they're going a little bit more towards the consumer category here rather than the enterprise segment. Is that right? Yeah, it's kind of their first product geared towards more casual users versus developers. And it's something that the cursor CEO has kind of talked about as, you know, a new category for them that they plan to expand into. And do you think that this is a smart strategy for them going into consumer? I mean, Cursor's CEO certainly does. He has good staff. He's told staff that, like, customers are requesting this, which is interesting.