Limitless Podcast - THIS WEEK IN AI: Pope Approves Anthropic, Comparing AI Plans, Starcloud SpaceX Partnership

Episode Date: May 29, 2026

We need to discuss Meta’s new subscription plans for Instagram, Facebook, WhatsApp, and AI. Then we'll compare them with offerings from Google, Claude, ChatGPT, and X. Can Meta justify its AI spending? If not, what possible fallback strategies can they lean on?We also explore StarCloud’s planned orbital data center, Cognition’s funding round, a new agentic coding benchmark, Pope Leo’s AI manifesto, and Snowflake’s post-earnings stock jump.------🌌 LIMITLESS HQ ⬇️NEWSLETTER:    https://limitlessft.substack.com/FOLLOW ON X:   https://x.com/LimitlessFTSPOTIFY:             https://open.spotify.com/show/5oV29YUL8AzzwXkxEXlRMQAPPLE:                 https://podcasts.apple.com/us/podcast/limitless-podcast/id1813210890RSS FEED:           https://limitlessft.substack.com/------TIMESTAMPS0:00 Meta’s New AI Subscriptions2:51 Comparing Subscription Tiers9:23 Meta’s Revenue Pressure10:30 SpaceX Goes Orbital AI13:42 Cognition's Big Raise15:37 The AI Pope18:37 New Benchmarks, New Winners20:30 Snowflake’s SaaS Comeback------RESOURCESJosh: https://x.com/JoshKaleEjaaz: https://x.com/cryptopunk7213------Not financial or tax advice. See our investment disclosures here:https://www.bankless.com/disclosures⁠

Everyone needs to stop forgetting about meta, the social media company turned AI company, because this week they have released quite a few new plans and options that I think are going to be interesting to someone. If you're watching this, chances are you will be affected by these new subscriptions. What we're going to do is not only share the new meta subscriptions, but also compare them to all the existing subscriptions that you can go and pay for, whether it be for social media platforms like X or through AI like Gemini, Claude, Anthropic, there is now a full spread of all the different types of subscriptions you could have to different models,

and we're going to walk through them systematically starting with meta, who started on the social front this week, offering $4 plans for Instagram, Facebook, and WhatsApp. Yeah, exactly. So Meta's getting into the subscription game, which they're probably following in the lead of Anthropic and Open Air. They're seeing these annual recurring revenue numbers just kind of skyrocket, and they're like,

I want some of that as well. Also, a bit of important context here, metas now spent around $30 billion between AI CapEx GPUs and like acquiring talent and trying to build their own AI products. But I mean, Josh, I think you share the same opinion. Which of the meta AI products have you actually used? This is their kind of like foray to try and get people to use it. So it comes in the form of a few different subscriptions.

Now, the consumer ones look like Instagram Plus for not $4, but $3.99 a month. You've got Facebook Plus at $3.99 a month. And you've got WhatsApp plus at $2.99 a month. Now, if you're wondering, like, why anyone would be. using these kind of like individual apps and individual subscriptions. It's because Instagram kind of caters to towards a content creator focused type of

platform. Facebook plus a lot of enterprises and business run their business on their business via ads. At the ad business for meta alone is absolutely huge and they're actually growing at a faster rate than Google did at the equivalent time that they existed in their life cycle right now. And then you've got WhatsApp as well, which handles a lot

of customer support for a lot of these enterprises. So you can think about one particular user, not just subscribing to one of these subscriptions, but many. And then collectively, there's this new enterprise plan. I think it's on the Meta-One AI subscription plan, which we'll get into it in a second, which is kind of like this collective suite of access.

Now, the thing that jumped out to me the most here is, honestly, how cheap the accessibility is here. If you are a content creator, the chances are it's a very competitive game, and you're not really earning as much money as some of these Fortune 500 companies. Now, if you want to get access to some, maybe not frontier frontier,

intelligence, but a model that's good enough to do the thing that you're hyper-focused on, such as advertising or creating content or using the right SEO timed words, meta is in the attention game. That is what all their models have been trained on, using their own data. They've been in this game for over a decade, and they know how to capture people's attention. That's very useful for the particular type of users that they're going after. But I wonder how this compares directly to some of the other major frontier labs like Google or Anthropic. Yeah, a helpful way of looking at Metis plans is there's three different types of them. One is the one that we just mentioned for $4 a month. You get access to Facebook and

Instagram Plus, which gives access to extra features like profile customization, super reactions, story insights, among other things. This is still very new. We're still figuring out what exactly this includes. Second to this, sitting on top of it, are their standard plus premium and max plans. Those are if you are an actual creator on the platform. That gives you the verified badge. If you see people with blue check marks, not on a platform like X, they are paying for these plans. that includes lots of things noteworthy. It doesn't include the skipping of ads. No matter what, you cannot escape ads it appears. And then third to this, we have Meta's AI plan. And Meta's AI plan is the one that we're going to sit on. We're going to spend some time on, and we're going to

start comparing to the other companies, because it's starting pretty cheap. Meta's AI subscription service starts at $8 a month. What does that get you? We're not entirely short just yet. It's called meta1 plus, and it will give you access to presumably the latest in meta's AI tools. What are AI tools look like. We're not entirely sure. This is all very new and novel. But the second thing is the $20 a month plan. And the $20 month plan is seeming to be the one that wants to compete head to head with the other companies like Claude and ChatGPT and Gemini. So how does it compare is what we're going to get into here? I think a good place to start is looking at their creative plans and pricing that they have for their existing subscriptions. Now, this isn't the latest descriptions that we just spoke about, but it kind of

implies what the other kind of new subscription types will include. Now, I've noticed that across plus, premium and max, you have this thing being brandished right in front of you, which is protect your brand. I think a lot of meta's users who are business users who care about the kind of social image and branding online across all their different platforms, get a lot of impersonators. So this is something that like I think AI is going to result in a proliferation of. It's just kind of like fake content, people that copy your content, copyright issues, etc. This kind of like collectively protects you against that. I'm wondering how that actually works,

but it seems like there's some form of agent interaction, whether they're human or digital. There's also maximized discovery. I think this is really where meta kind of shines compared to like the Google AI models, the Anthropic AI models, and open air themselves. Meta really has been focused on the search engine optimization role of things. Now, in a world of AI, you can think that the search engine kind of like evolves quite a bit.

Google's been speaking about this quite a bit, which is the AI model itself or the AI agent ends up becoming the customer and you need to cater to an AI agent versus a human. That SEO role looks very different and Meta's trying to pioneer what that model looks like. The first model that they released a few months ago, I believe, is New Spark and it excels at this type of work, whereas compared to like regular ILO-LOM work, it kind of sucks. Now, I want to look at like a direct comparison between Google Meta and the other likes, and we have this really cool artifact here. Yes, so thank you, Claude, for creating this artifact. We are clearly paying for the

Claude plan, or at least I am to say the least. But this is a really helpful table that compares the different types of offerings from all these companies. We have meta, Google, Cloud, ChatGPT, and X slash GROC. The angel level pricing is something that is used by almost all companies except for Cloud at $8 per month. You can get access to Meta1, Google's AI plan, ChatGPT, and X slash GROC. On X, that gives you the blue check mark plus access to GROC. ChatGPT gives you the Go plan. So you get ads in side of your offerings that when you type into your LLM. Cloud noteworthy doesn't do ads so they have no plan at this level. Google Plus is AI supported and then the same with meta one. Now standard is

$20 a month plan. X is noteworthy and the fact that they don't have one. They actually start at $30. But among the others, this is the point where I think a lot of people will pay the most attention because people are willing to pay $20 a month. And in most cases, it's very much worth it. If you are using a chat GPT plan, if you are using a Claude plan that's free, You're currently on an old, cheap model that doesn't really work nearly as well as the frontier models are. And if you want to judge the progress of AI models, you really need to be on this $20 month plan because it at least gives you access to the frontier models as they're being released, to Opus 4.7, to ChatGPT 5.5, and Gemini and soon Meta's new offerings. So if we're

comparing these, the standard level, I'm not sure who would really want to pay $20 for meta's offering when you can go and get chat GPT premium, Cloud Premium, Google Premium, even at the same $20 price point. That's a bit difficult. Then these go up and really just the higher tiers of this are more tokens. So there's not a huge difference in terms of feature set. It's mostly just the ability to use it more productively. So as a starting point, the $20 month plans are all very competitive, except for the fact that two of these are frontier models and the others are not. So if you're looking for Frontier-level intelligence, top intelligence per token, it's really cloud in chat GPT. Google has a lot of fun tools if that's where you like to spend your time. But aside from

that, meta and grok's offerings leave a lot to be desired. I'm kind of split on this, okay? So, like, on the one half, I agree with you. On the other half, I'm like, I do think meta's going off to a different type of market. Like, when I look at like a 799 a month plan, I instantly think of open air is chat GPT go plan. Do you remember? They launched that specifically just in India and I think it was like five bucks a month. That resulted in adding, I think it was like tens of millions of new users over the next couple of weeks since they launched it, right? And I think that Met is trying to do a similar thing here. They're going for a very specific archetype of individual, which is like, you're out in the content game. They understand that, like, you may not have an abundance of money yet, but they

want to kind of, like, lead you on that exponential. And they know that they have search and discovery kind of like nailed to the bone. And so they want to kind of like offer that or be the first major AI provider for that. Now, on the other hand of things, like, I can imagine, as I see all these new models released from Anthropic and Open AI, the next model release means that the previous models release is like even cheaper. In fact, the newer model releases end up being cheaper than the old model releases, which is kind of antithetical because you like get a smarter model, which is kind of cheaper than the previous model.

So it's like bang for your buck. I also think that with more GPUs and semiconductors, as that all scales, you end up just getting more tokens per watt, I believe it's performance per watt. It's just going to end up kind of driving lower. So I don't know about the staying. of meta's subscriptions here, their models do need to improve pretty rapidly. And this kind of leads me on to like an overarching theme that I have on meta, which is this is actually like good news for meta. And this is desperately needed by meta because they've blown around $30 billion

over the last 12 months acquiring talent and investing in this AI thing with nothing to show for it. They released a MewSpark model. It was kind of like whatever people have forgotten about it. People aren't really kind of using it aside from people who actually use meta's different platforms. but this could technically or theoretically on paper result in kind of like a gold mine for them. If you imagine like they can convert like 100 million users for like their $20 a month plan, that is reasonable income coming in. But I've also seen comments from Zuck earlier this week, which talks about if this doesn't play out this subscription thing, if their AI models aren't used by a lot of people, they're willing to go down the neocloud route,

which is what Elon Musk has done recently and partner with major AI labs to rent out their compute and data classes which have invested billions of dollars in. So I think Zuck is being pressured quite a lot by shareholders at this point. He's being forced to kind of show some kind of revenue. He fired or laid off 8,000 plus people over the last seven days. And that was like one of like, I think five cuts in the last couple of months. So Mehta's like backed into a corner right now and they need to kind of like show the goods or they need to pivot pretty aggressively. You mentioned the word neocloud, which is funny because I have some neocloud news.

This is, again, if you're listening to this, this is a Friday roundup. We're covering all the news. So we're going to speed run this. It's a very busy morning. Limitless HQ, so we're going to try to get through these as quick as we can. Starting with the StarCloud news, which I found to be really interesting, Elon and SpaceX have a new partner. They are working with, it's funny enough, Gavin Baker here is in the photo. We just record an episode with him yesterday about his portfolio.

Clearly, he is bullish on space. And the announcement is that StarCloud is building StarCloud 3 that is going to be launching on the new SpaceX Starship. This is Starship version 3. This is the test flight that they just had last week that we covered. And the idea is that that three times. spacecraft is designed to fit inside of the PES dispenser inside of Starship, the CEO, Philip Johnson, claims that StarCloud 3 would be the first orbital data center that's actually cost competitive with terrestrial at about five cents per kilowatt.

If commercial costs land near $500 per kilogram. So again, this is fully contingent on Starship being able to get it up into orbit. The idea is that commercial access should open up around 2028, 2029, given the amount of time that's going to take to get Starship online and then start launching these aircraft in space. But it's a really cool announcement where we have an actual client for SpaceX that is going to be launching serious payloads into space and fully pursuing the AI data centers and orbit plan. It's going to take a little while, but this is where we build the foundation. This is where companies are starting to plan these launches for the next couple of years.

We have another SpaceX AI alliance member, Josh. Over the last couple of weeks, you've got Elon partnering with Anthropic twice. You've got them partnering pretty aggressively with Curse, which is going to result in a $60 billion acquisition 30 days after the IPO, which is expected to be sometime next month. And now you've got this partnership with StarCloud. Now, what I love about this is if you think of SpaceX owning the highway or the toll booth into space, they're going to need the goods to be putting out into space, right?

So they've got Anthropic to kind of tentatively agree to say, like, yeah, we're going to actually use some of your space AI data centers. But also, you need the goods to get out there. Now, Elon Musk is investing a ton of money over the next five years into something called the TerraFab, which will end up becoming the largest chip fab in America if it plays out how he plans it to. And there's an interesting statistic from that goal, which is 70% of the chips that are produced are going to be radiation hardened and sent out into space.

So he's very much like committing his entire focus and goal into this. That's why the SpaceX IPO is valued. at expectedly a $2 trillion valuation, but he needs the goods to be able to do that. Now, I'm looking at this tweet up here saying that StarCloud recently or is in the process of raising $200 million at a $2 billion valuation, that is cheap.

If StarCloud can actually put out radiation-hardened GPUs, which, by the way, they have. There's currently, I think it's like a stack of like H-100s that are currently orbiting space that are training and inferencing on it, as Gavin mentioned in the previous tweet. What's stopping Elon from acquiring them and now having the actual goods to be able to do that?

Obviously, he's still reliant on Nvidia. He has a good partnership with them. But I just am extremely bullish about this. And I like that he's choosing to kind of partner with different firms versus trying to build everything himself. But moving on, there is another company that is raising a pretty decent valuation right now. Cognition, which is the, I believe they started Devon, yeah, two years ago, which was the OG AI agent. They raised over a billion dollars at $26 billion valuation.

Now, the coolest part about these guys is everyone thought that they were. were completely cooked after anthropic and open claw kind of like launched their like autonomous agents. But what I liked about cognition is they have a lot of staying power. And we've spoken about on previous episodes that there's this company called cursor. And everyone blamed them for being like kind of like or criticized them for being like a rapper, a thin wrapper around an AI model. But what ended up being the case was they had this sort of agent harness, this agent mode, which kind of like orchestrates how models work, uses the right models, uses the right prompts and it's an incredibly valuable mode when it comes to training a model. Cognition is probably

second in the ranking of having or owning this particular type of mode. And Josh, I know you wrote an entire essay about this recently. I'm a big fan of harnesses. I really like harnesses. If you haven't subscribed to a newsletter, I would highly recommend. I released an entire essay about harnesses early this week. It's basically all of the peripherals required to turn the brain of an LLM into a full body into something fully capable of doing agentic tasks that alarm. running. One thing I found funny from this announcement was the idea that Cognition uses now 90% of their own code. And if they're valued at a $26 billion valuation, clearly there's a lot of value there. So this is a continuation of the Rapper Wars. We now kind of have very clear winners

on who's winning the LLM wars, the large language models that sit on the frontier. How you package them, how you add a harness for them will be the differentiator. We saw Cursor with a $60 billion deal with SpaceX. Now we have Cognition raising another big round. The current war, that new frontier, being fought on the rapper level with the harnesses. Cognition's the newest entries in that. I'm pretty bullish, pretty excited on Cognition. Josh, on your bingo card for this week, did you expect the Pope to sign a partnership with a major frontier AILAP? A partnership. Tell me about this partnership. What do you mean partnership? Okay, so listen, we don't typically get into the swing of reviewing biblical

architecture here on this show, but we are going to today. Who got the mandate of heaven? Yeah, yeah, exactly. So Pope Leo. kind of blew the newsroom wide open this week, releasing his Magnifico Humannitas of His Holiness, Pope Leo, on safeguarding the human person in the time of artificial intelligence. Now, my initial reaction to this is I actually kind of love it because kind of like the religious pious leaders don't tend to kind of like technology a lot or get involved with it much.

The fact that he's written, I believe this is like a 50 page plus script on his kind of like thoughts of how this plays out, AI is going to evolve, how it kind of equates to like ethical use and religious scripts. He gets into a whole thing. Like there's this thing I was reading earlier on, but it's like between two biblical images and talks about the construction of the Tower of Babel and these two scenes within the Bible. And the main takeaway that I got from this entire thing is he's trying to be extremely thoughtful and careful about how these models are developed because he knows about the power

of influence that these models will have on kids, young adults, and everyone across all every single age in the future. And so he knows that it's going to guide a lot of education and learning about religious scripts and all these kinds of things. And so he wants to make sure that it's developed in the right way. Now, one of the biggest call to actions that he has out of this entire document and the preceding hearing or talk or discussion that they had after this was he wants there to be governance instilled into this.

Now, the governance, he suggests, comes from two particular parties. One being the national government that runs the entire country, respective country, where it's operating in, the models operating in. as well as the church or religious groups involved so that they can kind of mold the personality and ethics of this. Now, when I kind of like speak about this, I think of one particular AI lab, which is Anthropic, who have been speaking to religious leaders independently of their own accord

over the last, I believe, couple of months. And the Pope actually calls that Anthropic saying, like, we would like to walk together. So I thought that was a pretty interesting update. That's fun. It's nice when there's more awareness being raised around the idea. It's like this will bring a lot of new eyes to the world of AI. It will start a conversation that is going to need to happen.

It is going to bring awareness in a way that is constructive, not damaging, hopefully. And I think it's just, it's a fun, cool thing. And it's a testament to how large AI has gotten where this is now sitting at a global scale and it is top of mind. And I like the idea that all of these people are going to be working together to figure out the best way to navigate forward as a society. And that's pretty exciting. So glad I'm glad the Pope's getting on board. And hopefully a lot of people who, you know, respect and trust him will take AI a little bit more seriously and start considering this conversation. And then next on the docket, we have two more topics.

We're going to breeze through it. All right. Bang him out. Josh, something that you and I complain about a lot is when a new model releases, you have these fancy benchmarks, right? But we don't know what that means. And often these benchmarks are maxed out or gamified and doesn't actually tell us or portray what the model can actually do. say a model is super good at coding and then a software engineer actually uses it for their specific work, it's actually not good enough. Now, what a bunch of people ended up doing was they developed

this new benchmark called Deep SWE, and it focuses on agentic AI engineering specifically. So when you spin up multiple instances of Claude or ChatGBTGPT, for instance, to do a bunch of your coding work in parallel, how good is it actually? Now, the typical benchmark, and this is, by the way, like the Holy Bible, to use a pun here, of the Holy Grail of benchmarks, which is software engineering verified, which is what every AI lab kind of like shows off when they release a new model. Currently, Claude and ChatGBT at the top of this, but it's tested against pre-approved solutions and answers against a particular problem. So the model already knows the answer. And so it's not really authentic or genuine. What this benchmark ends up doing is,

providing not only new problems of which the models have no idea what the answer is, they're not given the answers beforehand, but also real long horizon tasks that go for between 7 to 12 hours. And so they're put to the test. And we have a new ranking over here. GBT 5.5 and GPT 5.4, their predecessor model, is number one and two of this new benchmark. You've got Claude Opus 4.7, just behind on 54%, and Sonnet 4.6 as well.

Gemini 3.5 Flash, even though it is their latest model, comes at a pretty poor 28%. Yikes. I do like the idea of more benchmarks. I think benchmarks are fun. They are important. They are oftentimes games. So if we can come up with ways in which we could organically test these, if you ask me, the way that I choose to do my benchmarks, it's purely vibes-based. I'll get in. I'll do the day-to-day task. You could very much feel a model if you use it every day. You could feel where it's better, feel where it's worse. I find that they're oftentimes spiky in the sense that they're very, very good at specific things and oftentimes a lot weaker in others. And that kind of spike

complex changes over time as these new models get released. But I'm glad we have a more formal way of doing this greater than vibes and greater than the things that have been game recently. Now, last thing on the agenda is Snowflake stock. The headline I'm seeing is Snowflake stock jumps 30% in a day. This was after the earnings report that just came out this week. And for those they don't know, Snowflake is a cloud compute providing platform. And clearly they've done something right because wow, this is a lot of revenue in one day. Yeah. So to give a bit of context here, there's something, there's a phenomenon happening called the SaaS apocalypse, software as a service apocalypse, where basically as Claude and ChatjibD end up upgrading their own models,

they become able to replace a bunch of subscription software tools and services because they could just vibe code it or rebuild it from scratch. And this has led to a massive decline. I'm talking, to hundreds of billions of dollars lost in the stock market for all these different SaaS products. I remember a time where Anthropic would release a plug-in update for, let's say, a security review tool, and all the cybersecurity stocks would absolutely dump, right? And so the point here is the trend has been kind of not only broken,

but Inverted. Snowflake, which is a major SaaS provider for a lot of enterprise companies, not only had their stock jump up 35%, but it's because they defined a moat that sat on top of Claude and ChatGPT. Now, in previous episodes, we've spoken about Anthropic and Open AI signing what's known as a joint venture with some major private equity firms. And the intention of this is to place their own engineers

that knows four deployed engineers into Fortune 100 companies so that they can kind of like design an architect-specific workflows, so AI-agentic workflows using Klaude or using chat GPT. Now, the only reason why they would be doing that is because their own generalized LLN isn't good enough. And so you need someone to kind of like guide this kind of orchestration. Snowflake is basically arguing their latest quarterly earnings

that they already have that data. They already have that design, and they're doing it themselves using a mixture of Claude and ChatGPT. And that's why the stock jumped up massively. It just shows that not only are SaaS companies here to stay, but they also might be having or creating a defensible mode. I feel like we need a dartboard that we can start throwing darts at to pick the companies that are going to go up 30%. Because every week, it seems like there is 10 new ones that are going absolutely nuclear. But that wraps up the episode today. Thank you all so much for watching. If you have made it through all of our previous episodes this week, you are now fully caught up.

You can go touch grass, enjoy your weekend. Have an amazing time, and we will be back, as always, with a new episode starting bright and early next week. Thank you guys so much for watching. Have an amazing weekend, and we'll see you next time.