Project

Future You: Explore Your Future Self with Personalized Generative AI

Copyright

Pat Pataranutaporn

Pat Pataranutaporn

Say hello to FutureYou!

*This project is oriented towards young people*

Inspired by studies showing how increased "future self-continuity"—the degree of continuity an individual feels with their future self—can positively influence saving behavior, academic performance, mental health, and subjective quality of life, our interactive, web-based chat platform seeks to similarly help users develop more long-term thinking and behavior by strengthening their relationship with a relatable yet virtual version of their future selves. The conversation with the future self is generated in real-time via a large language model that has been personalized based on a pre-intervention survey assessing user future goals and personal qualities. Have a chat with your future self:

If you feel like sharing your conversation online, please use the hashtag: #MITFutureYou

Copyright

MIT Media Lab

Q+A


How accurate of a representation is my future self?

Your future self is NOT a prophecy: it is a narrative of a potential version of the future generated from a large-language model that has been personalized on your initial survey responses. The future self cannot predict specific details about your future life–rather, it aims to paint a vivid and realistic picture of what your future life could be like.

Why doesn’t my future self seem like me?

The conversation with the future self is generated in real-time via a large language model that has been personalized based on a pre-intervention survey assessing user future goals and personal qualities. Therefore, it only reflects the data that you have put into the survey, and its responses are still subject to the training data it was originally trained on.

Will my data be saved?

Yes, it will be anonymously used for research purposes only.

Can I chat with my future self more than once?

Yes, you can 🙂.

Team

Pat Pataranutaporn, MIT

Kavin Winson, KBTG

Peggy Yin, Harvard University

Auttasak Lapapirojn, KBTG

Monchai Lertsutthiwong, KBTG

Hal Hershfield, UCLA

Pattie Maes, MIT

The Project is supported by KBTG