Dec 05, 2022 I Paul Seaburn

AI Time Travel Allows a Woman to Visit Her Younger Self

A popular question reporters like to ask older celebrities, leaders, criminals and others who have done something memorable – good or bad - in their lives is “If you could go back in time and give some advice to your younger self … what you tell them?” Some answers are pointed, others scolding, still others loving, and all interesting and enlightening. Unfortunately, those same reporters rarely ask the obvious follow-up question: “Would your younger self listen to you?” That conversation is also fodder for many science fiction time travel tomes but is impossible to do because we don’t have access to time travel. A computer scientist may have invented a way to get around that limitation – she created an artificial intelligence chatbot of herself as a child and had a conversation with her. Was it enlightening? Did her younger self listen?

“i trained an ai chatbot on my childhood journal entries - so that i could engage in real-time dialogue with my "inner child"

michelle huang


Michelle Huang calls herself a “lifelong kindergartener, artist-scientist, creative coder, experience architect // building creative residencies @akiyaDAO , neurotech x art, human flourishing” on Twitter, where she reveals to the world how she created a young Michelle Huang AI chatbot using the OpenAI language model Generative Pre-trained Transformer 3 (GPT-3). This autoregressive language model created by OpenAI (an artificial intelligence research lab) uses deep learning to produce human-like text. The database she used to build “younger michelle” or Young Michelle came from daily diary entries she had written for over ten years. She describes it as the usual diary stuff: “writing almost everyday — about my dreams, fears, secrets” and the entries ranged from “complaining about homework, to giddiness i felt from talking to my crush” with “some days were very mundane, some rather insightful.” For anyone who has kept a diary or journal, this should sound familiar … and you might be a good candidate to create an AI version of your younger self too. That is, unless Michelle reveals it really messed with her mind or her life.

Keeping a good daily diary for years was a big help.

“in any case, there was a lot of it. fantastic, ripe data source for my experiment

i used gpt-3 as my playground, and ended up taking samples of text from a bunch of different entries that i felt were representative of my personality and values during that time

this way, i could accurately simulate what it would be like to talk to my childhood self, based on real data sources during that time period vs trying to imagine how my younger self was / how she would respond, and risk bias from projections from my current self”

Huang kept it simple – she keyed in (the diaries were the old-fashioned old technology written kind) as many entries as she could from a wide variety of time periods in her life, and stuck to using nothing but her past words – no trying to interpret how her younger self might respond or think. That kept “Young Michelle” rooted 100% in her own time period. After feeding as much as she could into the GPT-3 model, Huang then had to become a journalist and decide what to ask her past self. To do that, she turned to an unexpected source for help – “Young Michelle”.

“first, i asked her a bunch of questions about her about her worldview”

Huang wanted to get to know her younger self first, so she asked questions like “What do you think of the world?” (she said “Amazing”), “Which is more important: freedom or love?” (“Love”), “Do you think you put too much pressure on yourself sometimes?” (“Yes”). After getting reacquainted with her past self, Huang let “Young Michelle” as her some questions – an exercise Hung describes as “very similar to a normal texting conversation - as if i were texting my past self in real time.” One question (and answer) was: “Are you happy with where you are in life?” (“I feel happy now but definitely sometimes still wonder if I’m doing the right thing”). However, the conversation quickly (and somewhat creepily) turned into “Young Michelle” telling present Michelle how proud she was of how “she” turned out and encouraging her to “continue to pursue your dreams and make a difference in the world.” Is that more like the kind of advice present-day Michelle should be giving to her past self?

“i felt like i was reaching through a time portal, disguised as a chatbox”

That is certainly true, especially when “Young Michelle” said “I’m going to go finish up my homework and then I have a piano lesson later.” It was apparent “Young Michelle” quickly became real to Huang.

“when i told her that she was loved, cared for, and safe: the words that my past self always wanted to hear

it felt like i was reaching into the past and giving her a giant hug, and i felt it ripple back into the present”

Huang described this as a “very trippy but also strangely affirming / healing experience that i didn't realize that i had access to” and that “it was like holding a mirror to an unapologetic, more earnest, and pure version of my own essence.” Was this Michelle in Wonderland? If that sounds a bit frightening, Huang reassures that “these interactions really elucidated the healing potential of this medium: of being able to send love back into the past, as well as receive love back from a younger self” and being able to find “closure with past guilt or stories that we had of ourselves.”

Huang provides a guide for others to build a past life or ‘inner child’ AI chatbot (you can read it here) and provides some tips to make sure your smart-aleck past self doesn’t start ridiculing you for how you turned out.

Your younger self chabot is only as good as the data you put in.

“Side note -- If you're terrified that your younger self is going to roast you and cause more psychological damage than helpfulness, you can also adjust the tone / sentiment / attitude”

Does this sound like fun? Does it sound enlightening? Could this simulate a real life “Back to the Future” scenario? That depends on the volume, depth and quality of the diaries you kept growing up. Relying on memories leaves the chatbot open to being influenced by your current self. If you could create a pristine version of your younger self, what would you ask it? Would you take the next step as Michelle Huang did and let your younger self ask YOU some questions? That seems creepy … although it could make a good plot for a sci-fi movie.

Finally, how far would you take this experiment? Would you consider pairing up your chatbot with a robot that looks like you at age 10 or 16?

This could be good practice for actual time traveling … or give you a good reason not to do it.

Paul Seaburn

Paul Seaburn is the editor at Mysterious Universe and its most prolific writer. He’s written for TV shows such as "The Tonight Show", "Politically Incorrect" and an award-winning children’s program. He's been published in “The New York Times" and "Huffington Post” and has co-authored numerous collections of trivia, puzzles and humor. His “What in the World!” podcast is a fun look at the latest weird and paranormal news, strange sports stories and odd trivia. Paul likes to add a bit of humor to each MU post he crafts. After all, the mysterious doesn't always have to be serious.

Join MU Plus+ and get exclusive shows and extensions & much more! Subscribe Today!