The Future of Decentralized Social Networks: How to Solve the Human Identity Problem

Author: Paul Veradittakit, Partner, Pantera Capital; Translation: Jinse Finance xiaozou

In 2017, a group of researchers from the MIT Media Lab claimed in Wired magazine that decentralized social networks "will never work." In their article, they cite three impossible challenges: (1) acquiring (and retaining) users from the ground up; (2) the (mis)handling of users’ personal information; (3) lucrative targeting The user's user ad. For all three, they argue, the economies of scale of existing tech giants such as Facebook, Twitter and Google are too large to make room for any significant competition.

Five years later, what was once considered "impossible" no longer seems so out of reach, and the way we conceptualize social media networks seems to be starting to shift. In this article, we will investigate how to solve these "old" problems using new ideas of Decentralized Social (DeSo).

1**、User identity issues of social media**

Modern social media faces unique bot problems. While social media platforms have an obligation to uphold free speech, the issue gets tricky when the "users" aren't actually real users, but bots. And, as it turns out, bots can have a major impact on public discourse, from alleged manipulation of the U.S. presidential election to shaping COVID-19 sentiment. Especially with the emphasis on anonymity, security, and privacy, any decentralized social media platform inevitably inherits the “bot problem”—essentially, how do you convince people that accounts on your platform are Real people instead of robots, especially in the age of advanced artificial intelligence?

Traditional KYC protocols are a naive approach that immediately suffers from privacy issues - the other side of the coin. How (and why) you should trust a social media platform to have our trove of sensitive data (from identities to private messages and financial transactions) that can recreate a person's entire personal and social life and professional life.

The question of "user identity" is thus a tension between confirming the identity of a user as a real person and guaranteeing the privacy of personal data. In this article, we will explore two approaches to this problem, one is biometric authentication (using zero-knowledge proofs), and the other is social vouching.

2**, Worldcoin and biometric authentication**

In the field of "proof of personality" problems, Worldcoin is arguably the most high-profile and controversial project. Besides having the backing of OpenAI’s well-known CEO Sam Altman, Worldcoin’s solution to the “proof of personality” problem is simple: use an iris scan to create biometric proof that you’re human (since robots don’t yet have retinas), and earn authentication tokens from it. In terms of data privacy, Worldcoin claims to use zero-knowledge proofs to ensure that acquired biometric data is stored securely.

The theory behind Worldcoin is that as artificial intelligence plays an increasingly important role in society, there needs to be a way to distinguish humans from robots, and key ones that are private and decentralized. By using the Worldcoin orb's iris scanning feature, people can be issued a "digital passport-like" World ID, making it possible for recipients to qualify for the cryptocurrency-based Universal Basic Income (UBI) mechanism and participate in global democratic governance new mechanism. Essentially, this World ID is intended to serve as a social primitive to guide future digital social networks.

In its documentation, Worldcoin emphasizes how its solutions put privacy first. For example, it claims to delete Orb-collected images, store only hashes of users' irises, and run zero-knowledge proofs (zk-SNARKs) to share personality-proof information without revealing any personal data. Although in the current promotion stage, these hash values are stored in a centralized database, in the long run, the team is committed to storing these iris hash data on-chain after the hash algorithm is fully mature.

But despite these privacy statements, there is still much debate about the true guarantees of privacy, security, and fairness. For example, it has been claimed that Worldcoin operator credentials were stolen and World IDs were sold on the digital black market so users could earn Worldcoin tokens without an iris scan. There is also the overall fairness issue, with MIT Technology Review magazine publishing a scathing article in April 2022 discussing how it deceived, manipulated and exploited nearly 500,000 users (mainly in developing countries) during its beta phase , and even called it a kind of "crypto-colonialism." In fact, as of August 2, 2023, Kenya, Worldcoin's former largest gathering place, has banned Worldcoin scans due to security, privacy, and financial concerns.

In addition to the controversy surrounding these specific projects, there are broader concerns about Worldcoin's general approach to biometric authentication via specialized hardware. Because the Orb is essentially a hardware device, even if the Worldcoin software were perfect, there is no guarantee that there will be no hardware backdoors that would allow Worldcoin (or other third-party manufacturers) to surreptitiously collect real biometric data from users, or insert fake configuration files into the system . To the skeptic, all of Worldcoin's privacy guarantees (zero-knowledge proofs, iris hashing, on-chain decentralization) seem like nothing more than enough irony to say, "Trust me bro , we are a trustless solution."

3**, Human ID and Social Voucher**

Another way to solve the problem of proof of personality is to use the social vouching method. Essentially, if verified human Alice, Bob, Charlie, David all "vouch" for Emily to be a verified human, then it's likely that Emily is also a human and not a robot. So the central question here is a game-theoretic design problem - how do we design incentives to maximize our ability to "validate humans".

Proof of Human Identity is one of the oldest and most important projects in this field. In order to "prove that you are human", you need to: (1) Submit your personal information, photos and videos, and 0.125 ETH deposit, (2) Have a registered human being vouch for you, (3) During the "challenge period" passed before. If someone challenges you during this time, the case will be referred to the Kleros decentralized court and the deposit will be at risk.

During the sponsorship process, users are first paired with sponsors through the sponsor form. After users are paired with their guarantors, they then video call to verify that the profile is a real match. Like the idea of Worldcoin, in the long run, the Proof of Humanity community also has a concept of a Universal Basic Income (UBI) available to those verified by Proof of Humanity.

Other similar projects that use social graphs for identity verification include BrightID's video call verification (where everyone can verify each other), Idena's continuous captcha creation and resolution game, and Circles' trust-based small groups.

Perhaps the biggest appeal of these social vouching-based platforms is that they seem less intrusive than Worldcoin, which actually requires you to scan your iris against a metal orb. Some of these methods, such as Idena's captcha "checkpoint mode", even seem to preserve a degree of anonymity without the need to share large amounts of personal data or third-party certificate authorities.

4**, the future of personality proof**

As artificial intelligence continues to develop and behave more and more like humans, designing innovative proof-of-personality mechanisms will become increasingly important, not only for universal basic income and other incentives discussed by many proof-of-personality projects, but more importantly As a way to better sanitize and regulate future social networks.

However, the process involves many trade-offs, from data privacy to the intrusiveness of the process, to determining the validity of personhood, and is one of the famous "cryptocurrency puzzles." As V God himself said, no single form of proof of personality seems to be ideal, and he suggested a possible hybrid path: use biometric-based methods at the beginning, but transition to a A more social graph-based approach.

Going forward, this is an area that needs to be more transparent about processes, code, and data. In short, there shouldn't be an ironic paradox where users need to "believe this is a trustless solution". Only in this way can we truly create a social network that lives up to the original vision of cryptocurrencies: decentralization and privacy.