Time machine: artificial intelligence made it possible to talk to yourself in the future

Chatbot encourages people today to think more about who they want to be tomorrow

Artificial intelligence researchers have created a «future own» chatbot that will inspire them to make wise life choices . MIT scientists hope connecting with your 60-year-old self will change your views on health, money and work.

Photo: pixabay.com

If your carefully laid out life plan has been thrown into disarray by spending time on the sofa, bingeing on fast food, drinking too much and not contributing to the company pension fund, it may be time to talk to your future self, writes The Guardian.

Without direct access to a time machine, researchers at the Massachusetts Institute of Technology (MIT) have created an artificial intelligence-powered chatbot that imitates an older user and shares observations and pearls of wisdom. The goal is to encourage people today to think more about who they want to be tomorrow.

Using a profile photo digitally aged to make younger users look like wrinkly, gray-haired seniors, the chatbot generates believable synthetic memories and uses the user's current aspirations to create stories about his successful life.

“The goal is to promote long-term change in thinking and behavior,” says Pat Pataranutaporn, a member of the Future You Project at the Massachusetts Institute of Technology Media Lab. “This could encourage people to make smarter choices in the present that contribute to their long-term well-being and life outcomes.”

In one conversation, a student who hoped to become a biology teacher asked a chatbot imitating a 60-year-old version of herself about the most rewarding moment of her career. The chatbot said he was a retired biology teacher from Boston and recalled a special moment when he helped a struggling student improve his grades. “It was so nice to see the student’s face light up with pride and achievement,” the chatbot said.

To interact with the chatbot, users are first asked to answer a series of questions about themselves, their friends and family, the past experiences that have shaped them, and the ideal life they envision for the future. They then upload a portrait image, which the program “ages” into. digitally to produce an image of the user aged 60.

The program then feeds the information from the user's responses into a large language model, which generates rich synthetic memories for the simulated older person. This ensures that when the chatbot answers questions, it is based on a coherent backstory.

The final part of the system is the chatbot itself, powered by OpenAI GPT3.5, which presents itself as a potential older version of the user who can talk about his life experiences.

Pataranutaporn had several conversations with his “future I”, but admitted that the most profound moment was when the chatbot reminded him that his parents wouldn't be around forever, so he should spend time with them while he could. “The session gave me a different perspective that still impresses me to this day,” he said.

Users are told that “I am in the future” is not a prediction, but rather a “potential future self” based on information people provide. They are asked to explore different futures by changing their answers to the questionnaire.

According to a scientific paper about the project, which has not yet been peer-reviewed, trials involving 344 volunteers found that conversations with the chatbot made people feel less anxious and more confident about their future selves. According to Pataranutaporn, this closer connection should lead to better life decisions — from focusing on specific goals and regular exercise to eating healthier and saving for the future.

Ivo Vlaev, professor of behavioral sciences at the University of Warwick, said people often have a hard time imagining themselves in the future, but this can lead to greater persistence in education, a healthier lifestyle and smarter financial planning.

He called the MIT project a “fascinating application.” principles of behavioral science. “He embodies the idea of ​​”nudge" – subtle interventions designed to direct behavior in a favorable direction, making the future self more noticeable and relevant to the present, – he said. “If implemented effectively, it could have a significant impact on the way people make decisions today in relation to their future well-being.”

“In practical terms, effectiveness is likely to depend on how well it can simulate meaningful and relevant conversations, – added the professor. – If users perceive the chatbot as authentic and insightful, this can significantly influence their behavior. However, if the interaction seems superficial or dodgy, the effect may be limited.