Social networking site that proves you're human: OpenAI steps into the redesign of "reality"

While generative AI is becoming a part of social infrastructure, social media is quietly reaching a critical point. In an age where AI can mass-produce content indistinguishable from human content—text, images, audio, and even video—social media platforms that cannot guarantee the identity of their content are no longer able to function as a vehicle for trust. Amid this, reports that OpenAI is developing its own social networking site based on biometric authentication are more than just rumors about a new business venture.

According to the report, OpenAI is planning a social networking site that only real users who have been verified as human can join, and is reportedly considering utilizing Apple's Face ID and World's iris authentication device, Orb. The development is being carried out by a small team of less than 10 people, and while leveraging the knowledge gained from ChatGPT and Sora, it aims to solve the bot problem that has long plagued X, Instagram, and TikTok.

The background to this initiative is the personal concerns of OpenAI CEO Sam Altman. Altman has been a long-time X user, but in recent years has publicly stated that AI-related posts and discussions feel more fake than before. AI's ability to easily generate plausible discourse has actually diluted the overall information density on social media, making it harder to identify trustworthy personalities.

Particularly serious is the rapid increase in fake content using AI-generated videos. Short videos are well-suited to algorithms, and provocative footage of unknown authenticity is more likely to spread. The current situation, in which an increasing number of operators are using fake videos that meticulously mimic human faces and voices to increase traffic and advertising revenue, is a structural problem that can no longer be dismissed as an isolated incident. The era in which video functioned as "evidence" is coming to an end, and trust in visual information itself is being shaken.

In this context, social media platforms based on biometric authentication could be one solution. If we can technically guarantee that a real person is behind each account, we can at least answer the question of "who is posting." Unlike phone numbers or email authentication, biometric information is difficult to replicate and is highly effective in preventing bots from creating large numbers of accounts.

However, there are still persistent concerns about biometric authentication. Iris data is immutable, posing a significant risk if it is leaked. From a privacy perspective, many voices have sounded the alarm about the centralized management of biometric information. Friction with regulatory authorities in various countries has also continued over the World Data Protection Act. If OpenAI were to move down this path, it may be necessary to design systems that "prove but do not disclose" biometric data, rather than simply storing it. This may require hashing or zero-knowledge proofs.

What is noteworthy here is the concept of self-sovereign IDs (DIDs) and soulbound tokens (SBTs), which have been developed in the Web 3.0 field. The idea of ​​making personality and achievements verifiable while retaining them in an individual capacity, rather than managing them centrally, is highly compatible with biometric social networking sites. OpenAI's vision is not simply a social networking site; it can also be seen as an experiment in addressing the fundamental question of "how do we prove we are human on the Internet?"

What I personally find interesting is that this movement could mark a turning point away from traffic-first thinking. We're moving away from an economy where fake videos and exaggerated statements garner attention, toward one where proven personalities and the accumulation of statements are valued. If engagement, such as likes and quotes, becomes a key metric, and social media returns to a space where "who is speaking" is emphasized, then this would be a healthy evolution.

In an age where AI can mimic human nature to the utmost, being human itself is becoming a rare attribute. OpenAI's biometric social networking initiative attempts to reflect this ironic reality. Can technology really rebuild trust? The answer lies ahead as this small team takes on the challenge.

Related Articles

The movement toward the integration of social media and finance accelerates.

Interactive Brokers Launches USDC Deposit Service