Meta's AI Hard Fork  ## Document Notes ## Highlights - **Personal VR Use Was Fleeting** - Jaeden Schafer recounts buying a Meta VR headset and briefly enjoying it over a holiday break. - He then stored it in a closet and rarely used it, illustrating VR's novelty value for many users. Transcript: Jaeden Schafer Before we do, if you want to be able to build tools, if you're not a developer, I'm not a developer. I built a platform called AIbox.ai where you can vibe code AI tools. You explain what you want to build and our builder will automatically link together multiple AI models, put in prompts and build a tool for you. So it automates the whole automation. You can then go use those and share those. If you want to check it out, it's a link in the description. It is AIbox.ai. All right, let's get into the podcast. The first thing I want to say is I don't think there's a lot of people crying right now over the metaverse, feeling like it is collapsing, essentially. Back in 2021, Facebook rebranded themselves as Meta, and they said there's going to be this new era of virtual reality. It was definitely a big shift from basically focusing themselves on social media. They're kind of like a new company. It was like a big new marketing thing. Part of the strategy hinged on they kind of had this belief that Gen Z was going to prefer to socialize inside of online games like Fortnite and Roblox instead of on sort of feed based apps Like Instagram. So of course, they did this big name change thing. And also this, the name change had another purpose, which basically was to distance them from, they had a lot of years of reputational damage tied to Facebook. So at that point, the brand was kind of, they had the whole Cambridge Analytica data scandal, and there's a bunch of like internal documents and whistleblower things that were basically Just making Facebook's brand not as good in how, you know, there's a bunch of whistleblowers and how it was harming teens and children. And there's all of these like congressional hearings over surveillance and misinformation and then allegations of them being a monopoly and then allegations of censorship. And then, you know, like there's just so much drama with the whole Facebook thing. So rebrand, call it meta. And they're just like, look, here's our new vision. We're going to build the metaverse. It's going to become the next great social platform. It's going to be, you know, a virtual world. Users are going to gather. They had Horizon Worlds was kind of the main app or hub where they did that. And they, you know, they said like, everyone's going to play and work and socialize using Meta's VR headsets. Okay, awesome. And to be fair, I purchased a Meta VR headset and played some games on there over a Christmas break, you know, between Christmas and New Year's. It was kind of fun. Then it sat in my closet and I never used it for a couple of years. So I think that was probably the experience a lot of people had. ([Time 0:01:00](https://share.snipd.com/snip/31352d6a-18a9-4676-8441-f68b3ddbd192)) - **VR Studios Shut Down Or Maintained** - Jaeden lists several VR studios affected by layoffs, including those behind Resident Evil 4 VR and Deadpool VR. - He notes Supernatural will stop producing new content and shift to maintenance mode after Meta's cuts. Transcript: Jaeden Schafer There's Twisted Pixel, which is kind of known for Marvel's Deadpool VR. There's Sanzaru, who's the creative of Asgard's Wrath. The VR fitness app Supernatural, which personally was one of my favorite when I had a VR headset. It was actually acquired by Meta in 2023 for about $400 million, but they're going to stop producing any new content and they're just going to shift into maintenance mode. So Camouflage, who's the studio behind Batman Arkham Shadow, has also seen a bunch of layoffs according to GeekWire. So I think The Verge also reported that Workrooms, which is Meta's attempt to bring, you know, VR into the workspace, also shutting down as well. ([Time 0:03:47](https://share.snipd.com/snip/57bdcea5-5176-48ef-a7e9-cf57e1e9a6c4)) - **The Metaverse Was A Colossal Spend** - Meta spent about $73 billion on Reality Labs, a sum equivalent to spending $1 million daily for 200 years. - That scale makes Meta's metaverse effort a massive strategic and financial gamble that largely failed to deliver. Transcript: Jaeden Schafer Altogether, Meta has put about $73 billion into the division. And I think to put that into perspective, you need to spend about a million dollars every single day for 200 years if you wanted to reach that amount of money. So that is not insignificant. That is pretty, basically pretty massively colossal of a failure. I think beyond the hype, if you're looking from like analysts and investors, a lot of the early versions of the metaverse were really bad products. The avatars were super lifeless. They were super awkward. They famously didn't have any legs. So there is one early Horizon World screenshot of Mark Zuckerberg's avatar, which was like went super viral because it just looked so horrible. He was saying like this was the future. And yeah, it was not great. So I think this kind of like building in the open strategy only works when consumers actually want the technology. If you're trying to build something and say, hey, everyone wants this, building the open is probably not the right direction. In VR's case, I think demand never really materialized at scale. Meta definitely quickly captured the majority of the VR headset maker market with Oculus and the Quest device. The sales then steadily kind of went down. Counterpoint Research reported that global VR headset shipments fell 12% year over year in 2024, making the third consecutive annual decline. Meta had about 77% of those shipments. ([Time 0:04:53](https://share.snipd.com/snip/86ad5038-a3f4-4ca4-a7ad-a38bdff81d16)) - **Design Flaws Hurt Early Adoption** - Early Horizon Worlds products suffered from poor design and lifeless avatars that failed to excite users. - Building publicly only works when consumers already want the technology, which VR lacked at scale. Transcript: Jaeden Schafer I think beyond the hype, if you're looking from like analysts and investors, a lot of the early versions of the metaverse were really bad products. The avatars were super lifeless. They were super awkward. They famously didn't have any legs. So there is one early Horizon World screenshot of Mark Zuckerberg's avatar, which was like went super viral because it just looked so horrible. He was saying like this was the future. And yeah, it was not great. So I think this kind of like building in the open strategy only works when consumers actually want the technology. If you're trying to build something and say, hey, everyone wants this, building the open is probably not the right direction. In VR's case, I think demand never really materialized at scale. ([Time 0:05:12](https://share.snipd.com/snip/dc3df057-aaf2-4ef6-8990-a3ce14f27243)) - **Big Platform Bets Met Harsh Reality** - Meta hoped the metaverse would escape Apple and Google's app distribution duopoly and create huge digital commerce. - Those forecasts were overly optimistic and the platform never reached expected user or developer scale. Transcript: Jaeden Schafer I think Meta's long term interest was never just hardware, they were a lot more focused on the economics of running their own platform, whether that was apps or games. And they're kind of looking at this as a way to escape from Apple and Google's grip on mobile app distribution. That was, I think, why they're putting so much money and really trying for this whole thing. During Facebook Connect back in 2021, Zuckerberg was openly criticizing the Apple Google duopoly. He said that building under the rules had been really quote unquote humbling and that higher fees were stifling innovation. He said that he wanted the metaverse to reach a billion users within a decade and generate hundreds of billions of dollars in digital commerce. There's a bunch of consulting firms, I think McKinsey was one of them. There's some banks, Citi, all of those were saying like, yeah, we expect to see the same kind of thing. They're kind of forecasting these multi trillion dollar metaverse economies by 2030. Seems like they were maybe ([Time 0:06:38](https://share.snipd.com/snip/833b678b-978f-490c-99f2-4e600028d456)) - **Cutting Developer Revenue Backfired** - Meta planned to take a 47.5% cut of digital sales on Horizon Worlds, combining a 30% platform fee plus 17% more. - High platform fees alienated developers and stifled the incentive to build on Meta's VR platform. Transcript: Jaeden Schafer He was talking about how he's going to extract revenue and all this kind of stuff. I think he did a little too soon. He probably should have waited and ramped it up. It's kind of funny because he was, you know, he criticized Apple and Google a lot and, you know, building under the duopoly is like so hard and I'm going to make a better thing for developers. So, but the interesting thing is that instead of giving a better deal than Apple or Google, Meta actually charged a lot more. So even before the VR, their VR headsets had really achieved a lot of scale, they said that they were going to take 47.5% of digital sales on Horizon Worlds, which is, you know, combining A 30% platform fee with an additional 17% cut. I think this is basically making, it made a lot of creators pretty mad, but it's like very uninspiring. ([Time 0:08:46](https://share.snipd.com/snip/6f9918a3-7434-45d6-9d2c-b54a3dc409e5)) - **Ray-Ban Glasses Are A Strategic Win** - Meta's Ray-Ban smart glasses became a surprising success, outselling traditional Ray-Bans in some stores. - This hardware and its AI features give Meta a clear path to pivot from VR to AI-driven wearables. Transcript: Jaeden Schafer The number one that I talk about all the time is Meta's Ray-Ban smart glasses. These have been incredibly popular. I know a ton of people that have them. And they give you essentially the ability to, without your hands, go record stuff. You can play music. They have a whole bunch of AI powered features. In some stores, they have outsold traditional Ray-Bans, apparently in 2024. Bloomberg recently said that Meta was considering doubling production to meet demand for these. And they also have a deal with Oakley's. So there's more glasses brands kind of coming on board with this. Right now, they have pushed further kind of into some of their AI powered wearables. And they're just basically to get the Meta AI out there more. There's obviously a lot more consumer interest, I think. Right now we have OpenAI and Amazon and a lot of other people exploring hardware. And I think, honestly, because they were in the metaverse, they spent, you know, over 70 billion dollars there. I think they were really well set up to kind of go and do these meta Ray-Bans. ([Time 0:10:27](https://share.snipd.com/snip/39e2c396-8efe-489b-a41b-c6384897810c)) - **Meta's Strategic Pivot Toward AI** - The metaverse lacked users, developers, and sustainable economics while AI is already a core product for Meta. - Shifting resources from the metaverse to AI and wearables is the logical next move for the company. Transcript: Jaeden Schafer So I think if you kind of look at everything together, the math doesn't really work. The metaverse wasn't getting enough users, developers, or any sort of sustainable economics. So AI, by contrast, is already becoming one of the core products on meta. And so I think for meta, killing the metaverse is kind of the obvious solution. The metaverse experiment, it feels like, is kind coming to an end i think the future at least for now is going to be in ai where meta is going to focus the most and i actually think that they're Going to be a really powerful player because of the meta ray-bans and because of some of the the programs they had in the past that have now sort of evolved to that as the final form factor So i think i would definitely not count meta out in this regard and i think it's their biggest competitive advantage when it comes to AI is kind of that hardware that has been successful. ([Time 0:12:04](https://share.snipd.com/snip/096acb18-b24b-42ea-8fba-8a06f2819d95))