While some panels at GDC 2022 have been deeply disappointing, the panel titled “A Brave New (Virtual) World: Ethics and Governance of XR and the Metaverse” was most definitely not among them. While it was a bit heavy at times and a few speakers occasionally sounded like they might wear tinfoil hats in their free time, I cam away from this one pleasantly surprised.
As we’ve noted, many of the metaverse panels this year have been more of an advertising platform for one game or another, but there was none of that at this panel. There was a wide variety of professions, educators, and designers across a spectrum of stances on what or how a metaverse may come to be, but all of it seemed fairly grounded in reality, though maybe not looking far enough into the future in terms of the use of VR and biometrics.
Who will control the metaverse?
We had the Video Games and XR Policy Affiliate at Berkman Klein Center at Harvard, Micaela Mantegna, moderating the talk, with panelists RealityPrime CTO At Large, Avi Bar-Zeev; Electronic Frontier Foundation Deputy Executive Director & General Counsel, Kurt Opsahl; University of Sussex Reader in Intellectual Property Law, Andres Guadamuz; and Dylan Urquidi, Unity Technologies & Unity Labs Research-Engineer & XR Designer. There were no PR people, no marketers, and no social media influencers.
Mantegna started off by noting that there’s still no unified agreement on exactly what “the metaverse” is, and I have to agree. Look no further than Roblox or Fortnite, which are both self-contained virtual worlds with no crossplay or cross-over with other externally created games. The “metaverses” that currently exist are still largely walled gardens; they’re not boundless, and there are “no connection points or ways of porting content” from one “metaverse” to another. More concretely, your Blizzard account, for all the little perks it may give you in Blizzard titles, may give Tracer in Overwatch and Heroes of the Storm through some perk, but no perk will make your Tracer appear in Call of Duty, much less Mario Kart.
This is also part of why if we had a metaverse at all, we’d likely have more than one, which seems fairly antithetical to the metaverse as originally described, but that hints at the problem. While Snowcrash’s metaverse seemed rather centralized, some people tossed the term around see it as a decentralized version of the internet, only furthering the term’s status as a meaningless PR buzzword. Perhaps this is why about a third of 2022 GDC respondents believed that the “metaverse concept will never deliver on its promise.”
Guadamuz in particular often argued as a skeptic, feeling that the metaverse as currently hyped is no different from any other virtual environment and will probably remain that way. Even if the metaverse were to become whatever’s after “the internet,” Guadamuz said it’ll largely be ruled by other virtual platforms, a decentralized web3 format, or a large open-user virtual platform. The last two would be quite different from what we currently have, but not necessarily better.
Urquidi noted that the transition to a “metaverse” might be hard due to copyright, which we’ll get into a bit later, but it’s largely why we haven’t really seen any decentralization, just more walled gardens. Without contracts and unified technical architecture, which we noted in our report on another panel isn’t really happening, we really shouldn’t expect any unified portal system for virtual world-hopping that would lead to anything like Ready Player One outside of big-budget Hollywood movies that pay to use various big-name IPs. And no, simply using using the appearance of various characters doesn’t make Fortnite a metaverse – it makes it yet another walled garden with deep pockets.
Bar-Zeev then argued that the internet was essentially made to be decentralized, but what people today are largely talking about is the democratization of the internet. Web3 as it’s largely been developing shows that our online structures are prone to decentralization. Look no further than NFT companies that form their own little clubs but don’t interact with other clubs. He also notes that it’s one thing when each person gets a vote, but when each token votes, it’s actually just another plutocracy where the wealthier you are, the more you get to call the shots, so really, we’re not even seeing democratization. Most of the current “clubs” are about the tokens, not people, so it’s more like various grift group carving out their own fiefdoms.
Guadamuz supported this concept about recentralization, while Opsahl supported decentralization; while he acknowledged the ongoing patterns of recentralization, he noted how centralized choices, which offer a strong sense of safety, have been abused. Metaverses may not be totally decentralized, but theoretically they offer opportunities to avoid the abuse that a fully centralized metaverse may allow.
That’s a nice thought, but once again it leaves us without a unifying metaverse. Bar-Zeev added that walled gardens may be natural in the creation process in that there may not be enough content outside of the garden, the walls should come down once the content outside that wall is higher, and Urquidi went further by saying that this is where open-source options should come in, but highly centralized platforms that don’t even allow a choice to peek outside are problematic if we want any kind of metaverse.
That led to discussion on why decentralization can be so difficult. As Opsahl argued, a centralized platform that holds onto a person’s content and social circles can be hard to break out of just in terms of user motivation. We’ve all been there with the avatar we spent an hour to make just right, only to find a flaw but not want to correct it, or worse, the character we’ve played for years that keeps us tied to an MMO that feels dying or dead. Yes, maybe a new MMO comes out, but none of your friends are going, and you’re so close to a milestone, so you just end up staying in your current game. We all jump virtual worlds at some point, but even with all the free trials out there, most of us won’t even give them the time of day unless we’re really needing a fresh start.
But we don’t exactly need a metaverse to do any of this because we already have it. Bar-Zeev says we need connections or portals between “worlds” that don’t also allow attacks between the worlds, both in terms of security but also feel. It’s less about bringing your fantasy dwarf to a sci-fi world (which, let’s be honest, most devs don’t truly want this) and more about the idea of having the freedom to take what you’ve done and take it elsewhere, like your phone data. Maybe your dwarf becomes something like a jawa in a sci-fi game, or your 500 game hours level-capped paladin grants you a similar level and class in your new game while keeping your name intact. The metaverse is supposed to give us some kind of bridge, but the issue is and most likely will remain one centered around money.
Paying for the product… or not
The 2022 GDC survey revealed that 70% of developers are completely uninterested in using NFTs – that is, 7% are interested, 1% are already doing it, and 21% are somewhat interested, but clearly the people pushing NFTs haven’t won over the people who are currently making virtual worlds.
Ever the doubter, Guadamuz notes that nothing would really change even if NFTs were to become popular. Despite what proponents often say, the developer will really own your NFT, not you. You may have the token, but legally you probably don’t “own” the asset you think you paid for. Maybe you would if you got a license the way Second Life gave them out to people who created stuff in that virtual world, but that’s not what we’re seeing in modern blockchain titles. Bar-Zeev worked on SL and noted that the team had used economists and created an economy of abundancy that promoted buying and using things, as opposed to one of scarcity where people simply buy things to hoard them and drive up their perceived value.
And that’s the problem: artificial scarcity. Opsahl notes that internet world can cut costs. As we’ve seen with NFT “art,” it’s easy to just copy and paste content. It costs nothing, especially NFT images. Rarely are you buying ownership of an image; it’s just a representation of some arbitrary code unless the seller actually gives you the copyright as well.
In other words, there’s no inherit legal ownership occurring in NFT purchases, especially when it comes to copyright. There’s already a specific process for that, and while we could theoretically make a digital version of the US Copyright Office for NFTs, that’s not a thing right now. Other countries, however, may not have a copyright registration system, and Guadamuz notes that this could make copyright issues easier or harder based on local laws. For example, in the UK, the copyright transfer needs to be in writing and signed with a signature. You could theoretically use a digital wallet to handle that, but he suspects NFT code would need to contain a kind of smart contract that doesn’t exist right now.
But some people still believe NFTs will evolve into something more than gifs attached to arbitrary code. Urquidi notes he’s working on an NFT project where the data are essentially built on, a long-touted idea without much evidence so far. The idea of the NFT recording your “experience” being “sold” elsewhere and continuing to gain “experience” to increase its perceived value is just that – a nice idea – and Urquidi believes that’s coming, but we’ve already seen that it may not be as near as some people claim.
That being said, Bar-Zeev added that it’d be nice to not only monetarily reward other people who may have added experience to a specific NFT if it were to make money but also give them credit if that NFT were to be merged into something, similar to how certain games allow players to build a virtual cog and get revenue/credit if that is used to make a clock by another player. This would allow for a kind of virtual resume, in which people could note they contributed to the experiential value of famous NFTs. Again, though, the problem is that that just isn’t happening, and there’s no evidence it’ll be happening any time soon, either.
Ethics, legality, and governance challenges
If it’s not obvious from the previous sections, there are tons of legal and ethical issues just surrounding who is making money and how things are being paid for, but that’s not even the tip of the iceberg. Opsahl notes that one of the challenges, at least in virtual reality, is that your device captures a whole lot of info about you, and all the data about you are being digitized and sent out to others, which can be used for everything from judgments by other players to figuring out how to advertise to you.
This is why Urquidi says we shouldn’t trust corporate entities or governments as these data can be abused without encryption tools (like zero-knowledge proofs and homomorphic encryption) and private biometrics, as these methods don’t store data that can be abused. But Urquidi goes further than just VR, claiming this is important not only for current technology but for future technology that may involve “implants” that record biometric data. While he doesn’t explicitly say so, my guess is that he’s referring to things like glucose monitoring implants or brain-computer interfaces and not, uh, surgically enlarged pectorals.
While it may sound a little sci-fi, the other panelists largely agreed with the old adage: If you’re not paying for the product, you are the product. As fun as VR may be, the data that may be recorded and used may be worth much more than whatever you pay for, if you even have to pay for it all. Abuse and exploitation are themes we see a lot of in this corner of tech, and simply convincing you to sign a waiver shouldn’t get companies off the hook. Bar-Zeev argues that written consent alone can’t be a get-out-of-jail-free card because even the scientists in these fields don’t totally understand what they may be handling. Opsahl adds that this is exactly why people don’t trust Facebook leaping into the metaverse (a level of distrust FB only reinforces when it releases pre-recorded “talks” on users and mishandling of data when other developers engage with its audience).
When asked by the audience how devs can avoid being part of the abuse, Bar-Zeev says those in the field already should make sure that any of that kind of biometric data are used only locally and never touch the cloud: don’t save it, don’t create meta-data from it, just delete it once you’ve used it in a way that helps the player. For example, maybe your data would be digitizing someone’s face for your game, but without encryption or basic security, a third party could siphon off those data to impersonate users in calls or video chats with their bank.
Then there’s the question of how to tackle the lack of knowledge of all these issues among lawmakers. Several speakers suggested the use of analogies to make people feel welcomed in the discussion, as opposed to using jargon and complex models that may alienate them. The more access lawmakers have to this kind of information, the better, Bar-Zeev says, noting how stories work better than simplifying the tech because they’re easier to share. Lawmakers should employ technologists, according to Opsahl, to help ease understanding. The metaverse may not be something everyone agrees on, but the issues surrounding it are becoming more and more obvious, and it’s clear that the technologies associated with the metaverse idea need to be addressed sooner rather than later.
