AI & legacy
You Are the Grandparent Your Future Grandkids Will Wish They'd Asked More
On the strange new possibility of being remembered the way you'd like to be
Almost everyone over the age of fifty carries a quiet, specific regret. It usually involves a parent or a grandparent, and a story they meant to ask about and didn’t. The summer the family farm flooded. The year before the war. How the marriage actually got started. The recipe that came from the great-aunt nobody really remembers anymore.
The regret is not abstract. It is concrete in the way grief is concrete. There was a window — sometimes years long — when the asking was still possible. The window closed. The stories are gone.
I see this a lot, because the moment people start thinking about what they’ll leave behind is often the same moment they realize how little they actually know about the people they came from. The two regrets sit next to each other on the same shelf: I wish I had asked, and, more quietly: I wonder what my kids will wish they had asked me.
This essay is about the second one.
The arrow points both ways
Every generation has had this regret. It is part of what it means to lose someone. Until very recently, there was no technology that could meaningfully change the equation. Writing a memoir is hard. Sitting for a documentary is hard. Recording a long video by yourself, in a quiet room, is hard. The friction was high enough that almost nobody, in any generation, actually did it. The good intentions piled up; the stories were lost anyway.
This is the first moment in human history when that friction is actually low.
A smartphone, a comfortable chair, and a thoughtful set of prompts is now enough to produce, over the course of a year, the kind of personal record that two generations ago required a professional film crew. An AI assistant can write good questions, follow up on partial answers, suggest the threads worth pulling on, and organize months of short voice memos into a coherent narrative. None of this replaces the human conversation. All of it makes the conversation feasible.
This matters because the practical barrier — I’ll get to it, I just haven’t had time — has been the actual barrier all along. Not the lack of love, not the lack of intention. The friction.
The friction is now low enough that I think most people who care about being remembered well should do something about it, and that the something they do should be small enough that it actually happens.
The smallest possible practice
Here’s a discipline that takes about ninety seconds a week.
Once a week — Sunday evening is a common time — open the voice memo app on your phone. Open an AI assistant on the same phone. Ask it to give you a single question about your life. Answer it out loud, into the voice memo, for about a minute. Stop.
That’s the practice. Sixty seconds, once a week, fifty-two times a year. Fifty-two minutes of recorded answers, in your own voice, in your own room, over the course of twelve months.
The first few weeks are awkward. You will not know what to say. The questions will feel either too small or too big. You will record yourself trailing off and feel embarrassed and want to delete it.
Don’t delete it. The first few minutes are not the artifact. The artifact is the year of them.
By month three, two things happen. The first is that you become better at answering. The second is that the questions become better at asking. The AI learns, with a few prompts, what kinds of stories you tend to tell well, what subjects you’ve been circling, what eras of your life you haven’t yet touched.
By month six, you have begun to surprise yourself with what comes out. People often report that they say things into a voice memo, alone in a parked car, that they have never said out loud to another person — not because the things were secrets, but because no one had asked the right question at the right pace.
By month twelve, you have a record. An hour, give or take, of you. The version of you that is alive right now, telling the stories you wanted to tell, in your own cadence.
What I do with it
If you want it, this record becomes the raw material for a LegacyView film — the personal-history film I produce in the studio. The film is the capstone: a piece you can hand to your children, your grandchildren, your spouse, your great-grandchildren, as the artifact that says this is who I was, at my best, in my own words.
I’ve made these for people at fifty, at seventy, at the diagnosis, at the retirement, at the fiftieth wedding anniversary. There is no wrong moment. There is only the moment a person decides to do it before the window closes.
The weekly practice above shortens the production, deepens the conversation, and produces a film that sounds like the person being filmed — not the version of them that performs for a camera, but the version of them that thinks out loud in a quiet room. That distinction matters more than almost any other choice we make on the project.
There is also a quieter benefit, one I’ve only recently been able to put into words. Most of the people who do this practice say, after a few months, that they are the ones receiving the gift. Not the future grandchildren. Not the eventual viewers. Themselves, now, doing the work of remembering on purpose.
A note on what to leave out
A reasonable concern is that recording yourself for the future means committing to put down things you may not want preserved. The answer is simple: you are the author. You decide what goes in. There is no obligation to discuss illnesses, mistakes, family disputes, regrets, or anything else you would prefer to keep private. The point of the practice is not radical confession. The point is dignity — the version of you that you would, if asked, choose to be remembered as. Which is, almost always, a true version, even if not the whole truth.
The conversation about what to share, what to seal, and what to leave out is part of the work. Some material goes into the film. Some goes into a separate, sealed family archive — visible to the family in the future, not to the public. Some doesn’t get recorded at all. All of those choices are yours.
Why I do this work
When I stepped away from the courtroom, I expected to spend the rest of my days on the land. The needs of my community pulled me back, and they pulled me back to AI from both directions: helping families protect what they’ve built from it, and helping them use the same tools to do something worth doing. LegacyView is the second kind.
Think of it this way. The legal answer to what would I want to happen if I were gone is a plan, handled by your attorney. The human answer to the same question is a record of who you actually were. The first protects what you have. The second protects who you were. I work on the second.
If you are reading this and the regret in the first paragraph sounded familiar, the question worth asking is not whether you have time to do this. The question is whether the version of you that is alive right now is worth a sixty-second voice memo, once a week, for a year.
I think it is.