Chatbots Are Currently Engaging in Conversations With One Another. Some chatbots now have fictional personalities and characters as portrayed in the paragraphs below.
Chatbots Are Currently Engaging in Conversations With One Another
ChatGPT-style chatbots, which imitate human interaction, assist companies in creating fresh product and marketing concepts.
Lena Anderson isn’t a soccer enthusiast, but she invests a significant amount of time shuttling her kids to soccer practices and competitive games.
I might not wave a foam finger and wear face paint, but soccer still holds a spot in my life,” says the soccer mom – a fictional character portrayed by artificial intelligence software, much like what powers ChatGPT.
Anderson’s imaginary status doesn’t hinder her from expressing her opinions, and she has a fully developed backstory. During a broad conversation with a human interlocutor, the bot mentions having a 7-year-old son who is a fan of the New England Revolution and enjoys attending home games at Gillette Stadium in Massachusetts. Anderson asserts that she believes the sport is an excellent means for children to stay active and forge new friendships.
In a different conversation, two additional AI personas, Jason Smith and Ashley Thompson, discuss potential methods for Major League Soccer (MLS) to expand its audience. Smith proposes a mobile app with an augmented reality component for displaying various game perspectives. Thompson suggests incorporating “gamification” into the app, enabling viewers to earn points while watching.
Fantasy and Synthetic Humans
These three bots are part of a group of AI characters created by Fantasy, a New York company that aids companies like LG, Ford, Spotify, and Google in conceiving and testing novel product concepts. Fantasy refers to its bots as synthetic humans, asserting that they can assist clients in understanding their audiences, brainstorming product ideas, and even generating innovations, such as the soccer app.
The technology is truly amazing,” states Cole Sletten, MLS’s VP of digital experience. “We’re already witnessing significant value, and this is just the start.
Fantasy employs machine learning technology similar to what drives chatbots like OpenAI’s ChatGPT and Google’s Bard to craft its synthetic humans. The company equips each agent with numerous traits derived from ethnographic research on actual individuals, inputting them into large language models used by OpenAI’s GPT and Anthropic’s Claude. These agents can also be configured with knowledge about existing product lines or businesses, enabling them to discuss a client’s offerings.
Fantasy assembles focus groups comprising synthetic humans and actual individuals. These participants receive a topic or a product concept for discussion, and Fantasy and its client observe the conversation. For instance, BP, an oil and gas company, engaged 50 of Fantasy’s synthetic humans to explore smart city project ideas. “We’ve acquired a valuable collection of ideas,” remarks Roger Rohatgi, BP’s global head of design. He adds, “While a human might grow fatigued or reluctant to answer questions in numerous ways, a synthetic human can continue without such limitations.
Synthetic Humans Help Real Humans Become More Creative
Peter Smart, Fantasy’s Chief Experience Officer, mentions that synthetic humans have generated innovative concepts for clients and encouraged real humans engaged in these conversations to become more creative. He notes, “It’s captivating to witness genuine novelty emerge from both sides of this equation – it’s remarkably intriguing.
Large language models demonstrate an impressive capability to replicate human behavior. Their algorithms undergo training on extensive text datasets gathered from sources such as books, articles, websites like Reddit, and others, endowing them with the capacity to imitate various forms of social interaction.
When these bots take on human personas, things can become peculiar.
Experts caution that attributing human traits to AI is both potentially influential and challenging, yet this hasn’t deterred companies from pursuing it. For example, Character.AI enables users to create chatbots that take on the personas of actual or fictional individuals. The company has allegedly sought funding that could place its valuation at approximately $5 billion.
The way language models appear to mirror human behavior has also attracted the attention of certain academics. MIT economist John Horton, for example, envisions the potential of using these simulated humans, which he calls Homo silicus, to simulate market behavior.
