Have you ever ever desired to travel through time to see what your future self may be like? Now, because of the ability of generative AI, you’ll be able to.
Researchers from MIT and elsewhere created a system that allows users to have a web-based, text-based conversation with an AI-generated simulation of their potential future self.
Dubbed Future You, the system is aimed toward helping young people improve their sense of future self-continuity, a psychological concept that describes how connected an individual feels with their future self.
Research has shown that a stronger sense of future self-continuity can positively influence how people make long-term decisions, from one’s likelihood to contribute to financial savings to their give attention to achieving academic success.
Future You utilizes a big language model that pulls on information provided by the user to generate a relatable, virtual version of the person at age 60. This simulated future self can answer questions on what someone’s life in the long run could possibly be like, in addition to offer advice or insights on the trail they may follow.
In an initial user study, the researchers found that after interacting with Future You for about half an hour, people reported decreased anxiety and felt a stronger sense of reference to their future selves.
“We don’t have an actual time machine yet, but AI generally is a form of virtual time machine. We are able to use this simulation to assist people think more about the implications of the alternatives they’re making today,” says Pat Pataranutaporn, a recent Media Lab doctoral graduate who’s actively developing a program to advance human-AI interaction research at MIT, and co-lead writer of a paper on Future You.
Pataranutaporn is joined on the paper by co-lead authors Kavin Winson, a researcher at KASIKORN Labs; and Peggy Yin, a Harvard University undergraduate; in addition to Auttasak Lapapirojn and Pichayoot Ouppaphan of KASIKORN Labs; and senior authors Monchai Lertsutthiwong, head of AI research on the KASIKORN Business-Technology Group; Pattie Maes, the Germeshausen Professor of Media, Arts, and Sciences and head of the Fluid Interfaces group at MIT, and Hal Hershfield, professor of selling, behavioral decision making, and psychology on the University of California at Los Angeles. The research might be presented on the IEEE Conference on Frontiers in Education.
A sensible simulation
Studies about conceptualizing one’s future self return to not less than the Nineteen Sixties. One early method aimed toward improving future self-continuity had people write letters to their future selves. More recently, researchers utilized virtual reality goggles to assist people visualize future versions of themselves.
But none of those methods were very interactive, limiting the impact they may have on a user.
With the arrival of generative AI and enormous language models like ChatGPT, the researchers saw a possibility to make a simulated future self that would discuss someone’s actual goals and aspirations during a standard conversation.
“The system makes the simulation very realistic. Future You is way more detailed than what an individual could provide you with by just imagining their future selves,” says Maes.
Users begin by answering a series of questions on their current lives, things which can be essential to them, and goals for the long run.
The AI system uses this information to create what the researchers call “future self memories” which offer a backstory the model pulls from when interacting with the user.
As an example, the chatbot could talk in regards to the highlights of somebody’s future profession or answer questions on how the user overcame a specific challenge. This is feasible because ChatGPT has been trained on extensive data involving people talking about their lives, careers, and good and bad experiences.
The user engages with the tool in two ways: through introspection, after they consider their life and goals as they construct their future selves, and retrospection, after they contemplate whether the simulation reflects who they see themselves becoming, says Yin.
“You may imagine Future You as a story search space. You’ve a probability to listen to how a few of your experiences, which should still be emotionally charged for you now, could possibly be metabolized over the course of time,” she says.
To assist people visualize their future selves, the system generates an age-progressed photo of the user. The chatbot can also be designed to supply vivid answers using phrases like “once I was your age,” so the simulation feels more like an actual future version of the person.
The power to take advice from an older version of oneself, slightly than a generic AI, can have a stronger positive impact on a user contemplating an uncertain future, Hershfield says.
“The interactive, vivid components of the platform give the user an anchor point and take something that would end in anxious rumination and make it more concrete and productive,” he adds.
But that realism could backfire if the simulation moves in a negative direction. To forestall this, they ensure Future You cautions users that it shows just one potential version of their future self, they usually have the agency to vary their lives. Providing alternate answers to the questionnaire yields a very different conversation.
“This will not be a prophesy, but slightly a possibility,” Pataranutaporn says.
Aiding self-development
To judge Future You, they conducted a user study with 344 individuals. Some users interacted with the system for 10-Half-hour, while others either interacted with a generic chatbot or only filled out surveys.
Participants who used Future You were capable of construct a more in-depth relationship with their ideal future selves, based on a statistical evaluation of their responses. These users also reported less anxiety in regards to the future after their interactions. As well as, Future You users said the conversation felt sincere and that their values and beliefs seemed consistent of their simulated future identities.
“This work forges a brand new path by taking a well-established psychological technique to visualise times to return — an avatar of the long run self — with leading edge AI. This is precisely the form of work academics ought to be specializing in as technology to construct virtual self models merges with large language models,” says Jeremy Bailenson, the Thomas More Storke Professor of Communication at Stanford University, who was not involved with this research.
Constructing off the outcomes of this initial user study, the researchers proceed to fine-tune the ways they establish context and prime users in order that they have conversations that help construct a stronger sense of future self-continuity.
“We would like to guide the user to speak about certain topics, slightly than asking their future selves who the subsequent president might be,” Pataranutaporn says.
Also they are adding safeguards to stop people from misusing the system. As an example, one could imagine an organization making a “future you” of a possible customer who achieves some great end result in life because they purchased a specific product.
Moving forward, the researchers want to check specific applications of Future You, perhaps by enabling people to explore different careers or visualize how their on a regular basis decisions could impact climate change.
Also they are gathering data from the Future You pilot to raised understand how people use the system.
“We don’t want people to turn out to be depending on this tool. Moderately, we hope it’s a meaningful experience that helps them see themselves and the world in another way, and helps with self-development,” Maes says.
The researchers acknowledge the support of Thanawit Prasongpongchai, a designer at KBTG and visiting scientist on the Media Lab.
