SocialAI: Navigating the Echo Chamber of AI-Generated Companions

To look at this article again, go to My Profile, and then click on Saved stories to see it

My Initial Gaze into the Abyss of AI with the SocialAI Application

Upon my inaugural engagement with SocialAI, I was convinced that the application was a piece of interactive art. This belief seemed the sole rational justification for my decision to voluntarily register for interactions with artificial intelligence entities going by names such as Blaze Fury and Trollington Nefarious, whose primary function appeared to be to provoke me.

Michael Sayman, the brain behind SocialAI, acknowledges that the app's concept might puzzle some users. In his recent unveiling of the application, he described it humorously as a personal social network inundated with countless AI-created comments that provide critiques, suggestions, and insights.

Indeed, SocialAI exists, assuming "exists" is appropriate for a digital realm where every individual you engage with is an artificial bot.

In the SocialAI setup, you're the sole genuine human participant. This latest iOS application enables you to share messages similar to how you would on platforms like Twitter or Threads. Right after you post, you'll notice an ellipsis, signaling that a response is being prepared, poised for a quick reply. Immediately following, numerous responses begin to show up beneath your original message, all crafted by AI entities. With the most recent update of the app, launched today, these artificial intelligences now also engage in conversations amongst themselves.

Upon registration, you're invited to select from a variety of AI persona archetypes: Are you interested in interactions with Admirers? Critics? Doubters? Eccentrics? Pessimists? Futurists? Geeks? Melodramatics? Progressives? Traditionalists? Introducing SocialAI, home to characters like Trollita Kafka, Vera D. Nothing, Sunshine Sparkle, Progressive Parker, Derek Dissent, and Professor Debaterson, all designed either to support your views or challenge them.

Image capture displaying the guidelines for configuring the Social AI application.

Is SocialAI truly shocking, or just an intensified reflection of today's social media landscape? The reality we often overlook is that our online feeds are already dominated by automated bots, shaped by complex algorithms, and fueled by AI-powered advertising mechanisms. We, as genuine users, contribute to this ecosystem by continuously providing new content for social applications, engaging with provocateurs, and making purchases. In return, we are entertained and sometimes experience a sense of connection with our friends and followers.

In his 1985 work, the infamous skeptic Neil Postman observed, “Anyone with a basic understanding of communications history is aware that each new thought technology comes with a compromise.” The compromise with social media in the era of AI is a portion of our human essence. SocialAI merely reduces the experience to sheer artificiality.

"Sayman notes that on many social media platforms, distinguishing between bots and actual individuals can be challenging. He expressed a desire to establish an environment where users can be certain they are interacting exclusively with AI, which he believes offers a sense of liberation."

One could argue that Sayman possesses a natural talent for app development. Growing up in Miami, Florida, amidst the financial downturn, Sayman rose to prominence by creating a collection of apps to aid his financially-struggling family, who were contemplating a return to Peru. His skillset eventually led him to roles in product development at major companies like Facebook, Google, and Roblox. Sayman's latest project, SocialAI, originates from his own startup, Friendly Apps, which has received venture capital funding.

His application serves as a prime example of design-centric approach over straightforward AI development. SocialAI doesn't function as a typical social platform, but rather as ChatGPT within the framework of a social sharing application. It strives to transform our engagement with generative AI technologies. Sayman suggests expanding beyond the traditional one-on-one chat interface by proposing a model where you can receive responses from multiple bots simultaneously.

During a recent Zoom call, he shared with me his perspective on generative AI, likening it to a smoothie in a world without cups. It's still possible to consume it using a bowl or a plate, though they're not the ideal containers. SocialAI, according to Sayman, might just be the perfect vessel.

Without hesitation, Sayman burst into laughter. "What a poor comparison," he remarked.

Sayman exudes charm and gives considerable thought to the integration of apps into our daily lives. Currently, he operates solo, utilizing OpenAI's technology as the backbone for SocialAI, while incorporating various custom AI models to enhance it. To avoid financial strain from the costs associated with using OpenAI, Sayman has implemented usage restrictions on the app, joking that without them, he'd face financial ruin in mere minutes. Moreover, he's still in the process of devising a profitable business model for SocialAI. Sayman acknowledges that he's not pioneering the AI-character app space; giants like Meta have already introduced AI characters into their platforms, and Character AI, which Google has recently taken a keen interest in, offers interactions with a vast array of AI personalities.

Sayman seems dismissive of the rivalry, explaining, "I don't envision my application as a space where users engage with figures they believe could be real. It's more about finding solutions to disputes or understanding whether your message could be offensive, and receiving input before sharing it elsewhere."

"Sayman mentioned that someone humorously suggested Elon Musk ought to try this out to preview his messages before sharing them on X."

I actually gave it a shot, throwing in some highly popular tweets from Elon Musk and the renowned Twitter personality Dril into my SocialAI stream. I posted a news article from WIRED; however, the link was inactive, given that SocialAI does not facilitate link sharing. (Besides, there's no one around to share it with.) I adapted the famous "Bean Dad" tweet, assuming the role of Bean Mom on SocialAI, encouraging my 9-year-old daughter to figure out how to open a can of beans on her own as a teaching moment. I shared political opinions. I inquired with my artificial SocialAI followers about who else I should be following.

Update crafted via Social Artificial Intelligence

The automated accounts complied and inundated my timeline with their remarks, reminiscent of Reply Guys but much more intense. However, their replies were devoid of substance or the complexity typically associated with human interaction. Primarily, I expressed to Sayman that everything seemed overly eerie, making it difficult for me to find any significance or value in the bots' contributions.

Sayman motivated me to create more content similar to the "Am I the Asshole" threads on Reddit: Am I at fault here? Do I owe my friend an apology? Is it justified to hold a grudge against my family indefinitely? According to Sayman, this is exactly what SocialAI is designed for. I gave it a shot. Initially, the comments from the SocialAI bot ignited my primal instincts and my sense of righteousness. However, Trollita Kafka quickly pointed out that, in reality, I was the one being unreasonable.

An evident shortfall in the roll-out of SocialAI by Sayman is the launch of a basic functional version without clear communication regarding crucial policies on privacy, content management, or the specifics of how SocialAI or OpenAI intends to utilize user data. This approach seems to follow the "move fast and break things" mantra. Although Sayman has stated that he does not use individuals' contributions for training his AI models, he highlights that users are still bound by the data usage policies of OpenAI, given his reliance on OpenAI's API for operation. Furthermore, there is no feature available for users to silence or block a bot that might be malfunctioning.

Fortunately, your feed remains confidential by default. There are no genuine followers to speak of. For instance, my supervisor at WIRED could sign up for SocialAI too, yet he wouldn't be able to track my activities or realize that I've reposted an Elon Musk tweet expressing a desire to repurchase Coca-Cola and reintroduce cocaine, similarly to how he's unable to monitor my ChatGPT profile and discover the topics of my inquiries.

On SocialAI, your interactions are exclusively with artificial intelligence; no real human contact is involved. This platform is designed to be a personal universe populated by AI entities that can either support or challenge you. While this concept may not appeal to everyone, it's a direction many are gravitating towards, possibly without even realizing it.

You May Also Enjoy …

Delivered to your inbox: A selection of the top and most peculiar tales from the WIRED collection.

Elon Musk poses a threat to national security

Interview: Meredith Whittaker Aims to Challenge Capitalist Ideals

What's the solution for tackling an issue such as Polestar?

Occasion: Be part of The Grand Interview happening on December 3rd in San Francisco.

Additional Content from WIRED

Critiques and Manuals

© 2024 Condé Nast. All rights reserved. Purchases made through our site may result in a commission for WIRED as part of our affiliate agreement with retail partners. Content from this site cannot be copied, shared, broadcast, stored, or utilized in any form without explicit consent from Condé Nast. Ad Choices

Choose a global website