Virtual influencers, or computer-generated characters that are just as active on social media as actual people, have become more prevalent in recent years. Many businesses are making investments in the “digital human economy,” a promising sector that Gartner projects will grow to $125 billion by 2035.
These days, it’s easy to create virtual influencers because of the obsession with AI image generators like Midjourney and Stable Diffusion. These tools allow anybody to create fake profiles with followers that interact with them like real online celebrities.
One such business, 1337 (pronounced Leet), is using generative AI to create a network of smaller content creators, or micro-influencers, driven by AI. These creators have a variety of backgrounds and hyper-individualized interests, and they aim to connect with individuals from niche communities such as gardening, emo music, vintage fashion, classical literature, and more.
The firm, which got its name from a hacker and gaming word from the 1980s, came out of stealth today with $4 million.
In addition to employing AI to generate these influencers, 1337 lets people recommend actions and words.
Co-founder and CEO Jenny Dearing told TechCrunch, “We have a unique opportunity to combine human interaction with early-stage AI today.” In a world where influencers are overly commercial or impersonal, 1337 presents various AI-driven entities that interact with people in novel and dynamic ways.
Users receive payment for their donations as well. The business presently charges a fixed rate, but it wouldn’t say how much it would set in public. Dearing stated, “Once our ‘Entities’ have followings, we will make this fee and bonus opportunities tied to the community’s engagement with content.”
1337 introduces 50 AI-driven influencers, dubbed “Entities,” each with unique abilities, characteristics, and passions. On the website of 1337, for example, Daria (she/her, they/them) is a 19-year-old, unapologetic music blogger who is passionate about “emo culture” and a “mental health advocate.” They even have a personal history; Daria, upon discovering their cousin’s vinyl record collection, was moved by the song’s “raw emotion” and took it upon herself to start a blog for other like-minded individuals with mental health issues.
Undoubtedly, each Entity has a complex life. They even discuss their worldview, preferred indoor and outdoor hangouts, and the design of their house. There are many distinct Entities to follow in 1337, each having a unique age, gender identity, nationality, and occupation. They also have public Spotify playlists, LinkedIn pages, and Instagram accounts. Instagram comments and direct messaging allow followers to communicate with the Entities.
Dearing continued, “We’re creating entities that evolve with their niche communities, adapting to the rapidly changing digital landscape where technology is evolving constantly. Our vision goes beyond mere chatbot interactions.” We sincerely believe that by doing this, we will change our understanding of social media participation and online interactions. We’re creating new ground in the process.
The entities are in beta testing. Their formal launch is scheduled for January 2024. The original 1337 team develops and designs entities with users and AI models, such as 1337’s proprietary solution, Midjourney (art), and OpenAI’s GPT-4 (for textual captions).
Co-founder Robin Raszka, who previously created the AI avatar business Alter before selling it to Google for $100 million in 2022, is also a member of the founding team with Dearing. Chief technological officer Jan Maly was formerly a machine learning engineer at software design business STRV. The team also includes a few critical strategic advisers, such as Bailey Richardson, the former Instagram community manager who presently oversees marketing and community at Substack.
Raszka told us, “We use large language models (LLMs) from open-source software (OSS) for multimedia content analysis, with an emphasis on images, and customized OSS LLMs integrated into our workflows.” “We had trouble keeping our virtual beings’ facial traits constant when creating produced material. Our solution was designed due to our quest for perfection; as a result, we can now add new entities and guarantee that their faces are constant throughout all photographs.
To co-create with an Entity, the user joins a Discord channel and responds to prompts to make four images of the Entity doing an action, posing in a specific spot, or going through an event. In addition, 1337 has a mode dubbed “Entity Point of View,” which allows users to see the world from the Entity’s perspective.
On behalf of the Entity, the moderation team accepts the post, adjusting the material according to predetermined standards such as their attitudes, actions, and voice. The originator of a post gets acknowledged in the caption upon approval.
Dearing stated, “Our creators—especially Gen-Zers—are exhibiting amazing engagement with the Entities and are producing and curating far more content every day than we can publish.”
At some point, 1337 will permit its “super creators” to start from zero and construct an entirely new Entity. The business will also provide a revenue model the following year in which producers who produce Entities receive a portion of sales as payment.
Dearing stated, “For instance, any brand collaboration would provide a revenue share model that heavily leans on supporting the creator versus 1337.” “We’ve also talked about creative business models that give creators ownership in the company, much like Substack has done.”
She said, “We are also looking at how we can help solopreneurs and nano/micro-influencers develop new strategies for interacting with their current audiences and communities to grow their enterprises.
The business informed us of some other intriguing innovations in the works, such as the ability for Entities to talk, which will enable them to create and host podcasts and films. In the first part of 2024, Audio for Entities will go live.
Credo Ventures, GFR Fund, Treble Capital, Roosh Ventures, Hugging Face CEO Clément Delangue, impact investor Natalia Vodianova, and model were among the investors in 1337’s fundraising round.
The funding’s primary goal is to support expanding its worldwide creator community. Hundreds of individuals have contributed to the Entities’ material thus far.
“Today, the domains of human creativity and connectedness are home to some of the most exciting uses of diffusion models and LLMs. This confirms our belief that social media platforms will be profoundly impacted by the next wave of AI, moving marketers, followers, and artists equally, according to Karolina Mrozkova, general partner and investor at Credo Ventures.
