I personally think The Oasis is totally gonna happen

Prediction: Ready Player One will be regarded as eerily prescient

Disclaimer: I’m a data scientist who isn’t an expert on any of the things I’m about to talk about. I’m happy to be proven wrong on anything I write, because being wrong is an opportunity to learn.

Virtual reality is a field that I think receives far too much doubt, particularly the term “metaverse”. I suspect this is probably mostly just fueled by negative sentiments associated with Facebook and its ilk. People understandably find the idea of Zuck having a total monopoly on human attention in his digital world a bit uncomfortable. What I see a lot, though, is this discomfort being used to rationalize why the concept of the metaverse itself is not feasible. I don’t think this is the right take.

I’m going to use “the metaverse” to refer to the Ready Player One kind of metaverse — a centralized blob of datacenters that hosts a massive, immersive open world game server which anyone can create a free account in and join at any time from anywhere in the world, so long as they have their VR gear. And yeah, I suspect it’s going to end up belonging to a corporation because capitalism is the only source of incentive strong enough to make something this huge happen, and capitalism doesn’t seem to be going anywhere any time soon.

What would be the requirements for this future to happen? I think they can be grouped into three categories: market requirements, software requirements, and hardware requirements. The first two categories are trivial, the third is more involved.

First, market requirements. I think a Ready Player One metaverse would catch on because every separate part of it already has. Social networks are insanely widespread and addictive. Billions of people use them. Real time MMO games being used as social networks have also been proven to be wildly popular, see: Fortnite, Minecraft, Roblox, etc.

Concepts like avatars and digital identities don’t need to “catch on” because they already have caught on. The metaverse would be a more immersive variant of the Internet culture we already have. People already make friends, make enemies, fall in love, and do business online all the time. Heck, VTubers are even a thing, and growing quicker than ever too!

The metaverse will simply be the next level of online escapism. Humans are addicted to escapist fantasies. The metaverse will cater those fantasies directly to our monkey brains. I expect a lot of people will spend most of their waking hours with a headset on. I think that’s bad, but I think it’s going to happen anyways.

Second, software requirements. I think these succumb to similar arguments: all the core ideas already exist, just at a smaller scale or without having been integrated with all the other pieces before. To be clear, I am not implying that it would be simple or easy to create and maintain the software that runs The Oasis — it would be massively difficult and require huge teams of people much smarter than me. But it’s doable. We’ve made live multiplayer on large scales work before. We’re making massive progress in real-time motion capture and retargeting of that capture to 3D characters (especially recently, thanks to ML). Managing absolutely colossal amounts of personal data is nothing new to the tech giants. Real time ray-tracing I’d argue is more of a hardware problem than a software one.

All the pieces exist, they just need a lot of refinement before they’re Oasis-tier.

A quick note about the business side of things here: The potential profit for the entity that first puts all the right pieces together in the right way is effectively unbounded. If done correctly, I think this would be the most influential software ever made. It would effectively harvest a significant fraction of all human attention worldwide to be directed towards whatever targets the entity desires — even more than traditional social networks harvest today. Manipulating people’s monkey brains is very, very lucrative. Lots of ad revenue.

Third, tech requirements. As a ballpark estimate, let’s say that the perfect VR headset needs an 8K HDR display in each eye with a combined full 180° field of view, running at 150Hz with sub-10ms latency controller and eye tracking, as well as real-time raytracing for rendering scenes in a physically-based manner. In addition, let’s assume it’s standard for everyone to own force-feedback haptic gloves and a full-body haptic suit, as well as an omnidirectional treadmill that can be sprinted on. I feel these specs are a fair baseline to work off of.

When you write that list out, it becomes more apparent just how close this all really is. The chip needed to power such a headset is only 4 or 5 doublings away from the comparatively puny Qualcomm chip that currently powers the Quest 2, the most popular consumer VR headset. Looking at the past few decades of computer trends, as well as the speed with which these things are currently improving, and 4 or 5 doublings is only maybe… 20 years away? And that’s starting with the Quest 2’s chip—if we’re starting from something like the M1, it’s even faster.

There are already omnidirectional treadmills and haptics suits and force-feedback gloves available to the public. At the moment they are expensive and sub-par in terms of performance, but they’re miles ahead of where they were 20 years ago.

20 years from now is 2042. Ernest Cline set his book Ready Player One in 2045. I predict that VR tech will not only be as powerful, but also as widespread as it is in that novel well before 2045. In fact, I’ll move my prediction even further back — I think it will be here before 2037, just 15 years from now.

I’m only 20 years old. Even the slowest computer I’ve used in my life would still probably be fast enough to have powered the entire Apollo program. But I do have some memories of computers freezing and taking a while to do simple things like photo and rich text editing. My little sister, who is 11, doesn’t have those memories (to the best of my knowledge). In the sub-decade gap between when I was born and when my sister was, the entire technological landscape had transformed to the point of unrecognizability.

I am young, yet just within my lifespan I have seen concepts go from being commonly called infeasible to being part of almost everyone’s day-to-day life (see: the smartphone, professional gaming, extremely deep neural networks).

When you look at it that way, worldwide VR adoption isn’t even that crazy, all things considered. In fact, it’s already almost within our reach. At this point, the metaverse isn’t a matter of if, but when. I’ll be honest, I’m a little scared — but I guess the whole point of this post is that being scared of something doesn’t mean you should be in denial about it.

The metaverse is coming, one way or another. I don’t know exactly what to expect and I’m terrified that it’s going to do horrible things to society. But I’m also hopeful that it could increase human empathy and connection, and who knows—maybe it will end up being a massive net positive? No one can know for sure. All we can do is prepare ourselves for it.

It’s kind of like the inevitable advent of AGI and the robot uprising, but that’s a personal crisis of mine that I’ll share some other time.

Thanks for reading,

Kai