AI agents are coming for your dating life

On a Monday afternoon in March, I watched a pixel art avatar roam the corridors of a virtual office campus looking for a mate. With dark brown hair and stubbled chin, the sprite was a representation of me—an AI agent instructed to talk to other people’s agents to see if we might vibe in real life. It jumped in his first interaction, “I’m Joel, by the way.”

Three London-based developers ran the simulation: Tomáš Hrdlička and siblings Joon Sang and Uri Lee. The thesis behind their project, Pixel Societies, is personalization AI agents Can help match real people with highly compatible colleagues, friends and even romantic partners.

Each agent runs on top of a customized version of a large language model, fed a mix of publicly available data about a person and any additional information they provide. The agents are supposed to function as high-fidelity digital twins, faithfully replicating a person’s mannerisms, speech, interests, and so on.

Let loose in simulation, my agent was more like a Hyde to my Jekyll. “I’m always looking for the less glamorous side of the story,” it told one agent, one of several journalistic clichés it uttered. “Hype is my daily bread,” it told another. It hallucinates a reporting trip to Sweden and later a non-existent story that said I was cooking up. It cuts several conversations short with the phrase, “Let’s skip the treats.”

Pixel Societies remains a mere proof-of-concept, and because I offered little personal data—the answers to a short personality quiz and links to my public social media—my agent was doomed to life as a walking, talking LinkedIn post. But the developers theorize that deeply trained agents can move through interactions at breakneck speed, gathering information their owners can use to find real companionship.

“As humans we only live one life. But what if we could live a million?” said Joon Sang Lee. “It will give us more latitude to experiment.”

“A Spicy Personality”

Pixel Societies was born in early March during a hackathon at University College London, hosted by Nvidia, HPE and Anthropic. Hrdlička and Joon Sang Lee are both members of Unicorn Mafia, an invitation-only group of developers who regularly compete in these kinds of engineering competitions. In this case, participants were told to simply build something simulation-related.

Over two days, together with Uri Lee, they developed Pixel Societies, using an image model to generate the sprites and coding automation tools to expand the code base. Then they simulated a mini-hackathon within the virtual world they created, populated with agents representing the other participants. Anthropic awarded the team an award for the best use of its agent tools.

I ran into Hrdlička a few weeks later at a workshop about OpenClawan agent personal assistant software that blew up in January and whose creator was later hired by OpenAI. (In his simulation, Joelbot interacted with agents belonging to other people at the OpenClaw workshop.) Pixel Societies draws great inspiration from OpenClaw, which broke ground with the invention of a “soul file” that informed each agent’s unique identity. “It’s like giving an agent a really spicy personality. That’s what we used to make the characters feel alive,” says Hrdlička.

Encouraged by the reception at the hackathon and among fellow Unicorn Mafia members, the trio aims to transform Pixel Societies into something that looks less like a closed-loop simulation and more like a social platform where agents interact freely and continuously, with the goal of sparking fruitful real-world relationships. They haven’t landed on a business model yet, but options include selling virtual items for avatar customization and credits for additional simulations.