Have you ever desired to travel through time to see what your future self might seem like? Thanks to the ability of generative AI, that is now possible.
Researchers from MIT and elsewhere have developed a system that permits users to have a text-based online conversation with an AI-generated simulation of their potential future selves.
Synchronized Future youThe aim of the system is to assist young people improve their perception future self-continuitya psychological concept that describes how connected an individual feels to their future self.
Research has shown that a stronger sense of future self-continuity can positively influence the way in which people make long-term decisions, from their likelihood of contributing to financial savings to their deal with academic success.
Future You uses a wealthy language model that relies on user-provided information to generate a relatable, virtual version of the person at age 60. This simulated future self can answer questions on what an individual's life is perhaps like in the longer term, offer advice or insight into the trail they could take.
In an initial user study, researchers found that after about half an hour of interacting with Future You, people reported reduced anxiety and felt a greater sense of reference to their future selves.
“We don’t have a real-time machine yet, but AI is usually a type of virtual time machine. “We can use this simulation to assist people think more about the implications of the selections they make today,” says Pat Pataranutaporn, a recent Media Lab graduate student who’s actively leading a program promoting human-AI interaction research MIT and Co. developed – lead writer by a Paper about 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 student; in addition to Auttasak Lapapirojn and Pichayoot Ouppaphan from KASIKORN Labs; and senior authors Monchai Lertsutthiwong, head of AI research at KASIKORN Business-Technology Group; Pattie Maes, Germeshausen Professor of Media, Arts and Sciences and leader of the Fluid Interfaces Group at MIT, and Hal Hershfield, professor of selling, behavioral decision making and psychology on the University of California, Los Angeles. The research will probably be presented on the IEEE Conference on Frontiers in Education.
A sensible simulation
Studies on the conceptualization of 1's future self date back to At least within the Sixties. An early approach to improving future self-continuity was for people to write down letters to their future selves. More recently, researchers have used Virtual reality glasses to assist people visualize future versions of themselves.
But none of those methods were very interactive, which limited the impact they might have on a user.
With the appearance of generative AI and huge language models like ChatGPT, researchers saw a possibility to create a simulated future self that would discuss an individual's actual goals and desires during a traditional conversation.
“The system makes the simulation very realistic. “Future You is rather more detailed than what an individual could provide you with if they only imagine their future self,” says Maes.
Users first answer a series of questions on their current life, things which are necessary to them, and goals for the longer term.
The AI system uses this information to create so-called “future self-memories,” which offer a backstory that the model draws from when interacting with the user.
For example, the chatbot could talk in regards to the highlights of an individual's future profession or answer questions on how the user overcame a specific challenge. This is feasible because ChatGPT is trained on extensive data where people speak about their lives, careers, and good and bad experiences.
The user engages with the tool in two ways: through introspection, as they reflect on their life and goals while constructing their future self, and thru retrospection, as they reflect on whether the simulation reflects who they themselves need to be says Yin.
“You can consider Future You as an area for locating stories. “It gives you a possibility to listen to how a few of your experiences, which could also be emotionally charged to you now, is perhaps metabolized over time,” she says.
To help people imagine their future selves, the system creates an age-matched photo of the user. The chatbot can also be designed to supply descriptive answers with phrases like “When I used to be your age,” making the simulation feel more like an actual future version of the person.
The ability to just accept advice from an older version of oneself, somewhat than from a generic AI, can have a stronger positive impact on a user contemplating an uncertain future, says Hershfield.
“The platform’s interactive, visual components give the user an anchor point and make something which may result in anxious rumination more concrete and productive,” he adds.
However, this realism could backfire if the simulation goes in a negative direction. To prevent this, they ensure that Future You warns users that it only shows one possible version of their future self and that they’ve the chance to vary their lives. Providing alternative answers to the questionnaire results in a totally different conversation.
“This will not be a prophecy, but somewhat a possibility,” says Pataranutaporn.
Supporting self-development
To evaluate Future You, they conducted a user study with 344 people. Some users interacted with the system for 10 to half-hour, while others either interacted with a generic chatbot or simply filled out surveys.
Participants who used Future You were capable of develop a better relationship with their ideal future self based on a statistical evaluation of their responses. These users also reported feeling less anxious in regards to the future after their interactions. Additionally, Future You users reported that the conversation felt sincere and that their values and beliefs seemed consistent across their simulated future identities.
“This work breaks recent ground through the use of a proven psychological technique to visualise times to come back – an avatar of the longer term self – using cutting-edge AI. This is precisely the type of work academics should deal with because the technology for constructing virtual self-models merges with large-scale language models,” says Jeremy Bailenson, Thomas More Storke Professor of Communication at Stanford University, who will not be involved on this research was.
Building on the outcomes of this initial user study, researchers proceed to refine the way in which they set context and highlight users to have conversations that help construct a stronger sense of future self-continuity.
“We need to guide the user to speak about specific topics as a substitute of asking themselves who the following president will probably be,” says Pataranutaporn.
They also add protections to forestall people from abusing the system. For example, one could imagine an organization creating the “future self” of a possible customer who achieves an amazing end result in life by purchasing a specific product.
In the longer term, researchers need to explore specific applications of Future You, perhaps by allowing people to explore different careers or imagine how their on a regular basis decisions might impact climate change.
They are also collecting 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. Rather, we hope it’s a meaningful experience that helps them see themselves and the world in a different way and aids in self-development,” says Maes.
