Artificial intelligence is increasingly permeating every aspect of our lives, from customer service chatbots to advanced machine learning systems that can mimic human behavior. Now, companies like Fantasy are pushing the boundaries of AI by creating synthetic humans – AI-powered bots that can interact with each other, carry on conversations, and even generate novel ideas. This development transforms how we think about artificial intelligence for businesses, academics, and everyday users.
“The technology is truly incredible. We’re already seeing huge value, and this is just the beginning.” – Cole Sletten, VP of digital experience at the MLS.
Fantasy: Creating Synthetic Humans
New York-based company Fantasy is at the forefront of creating synthetic humans. These bots, powered by machine learning technologies like OpenAI’s ChatGPT and Google’s Bard, are given dozens of characteristics drawn from ethnographic research on real people. Their capabilities extend beyond mere conversation – they can also be set up to know existing product lines or businesses, allowing them to converse intelligently about a client’s offerings.
AI-Driven Focus Groups and Idea Generation
Fantasy’s approach involves creating focus groups composed of synthetic humans and real people. Participants are given a topic or product idea to discuss, and the ensuing conversations provide valuable insights and new ideas. This method has proven to be a powerful tool for idea generation. For instance, oil and gas company BP asked a swarm of 50 of Fantasy’s synthetic humans to brainstorm ideas for smart city projects, yielding a rich trove of innovative concepts.
AI Bots and Human Behavior
Large language models, the backbone of synthetic humans, are remarkably adept at mirroring human behavior. These models are trained on vast amounts of text from books, articles, and websites, enabling them to mimic many types of social interaction. However, as these bots adopt human personas, exciting and sometimes odd scenarios can arise.
Potential Applications and Future Developments
The potential applications of these synthetic humans are vast. Academics like economist John Horton of MIT see potential in using these simulated humans, which he dubs “Homo siliceous,” to simulate market behavior. Moreover, software developers see exciting possibilities for these language-model-powered agents in testing software and developing video games.
However, there are also challenges and questions to be addressed. Although these models can reflect human behavior to some extent, they are less complex and intelligent than real people. They may tend to be more stereotypical and less varied than natural populations. How to make these models reflect reality more faithfully remains an open research question.
Despite these challenges, the emergence of AI-powered chatbots that can interact, converse, and generate ideas represents a significant step forward in artificial intelligence. As technology advances, we expect to see even more innovative applications of these synthetic humans.