Parenting in an AI world

I saw a post on LinkedIn that got me thinking about the world I'm parenting my six-year-old in. It's equal parts scary and exciting. Things are changing way too fast for me to truly keep up, and it will only accelerate from here. Yet, fundamentally, my job as a father isn't all that different than it was for my father—it's just inflected with today's technology.

I'm the same age as the personal computer. It puts me in a unique position to understand today's technology, since as I grew up, personal computing grew along with me. I'm part of the first generation of humans who truly had the opportunity to be digital natives. I've witnessed every era of the personal computing revolution, and I've adapted to each one.

On one hand, AI is the latest in a chain of technological quantum leaps—from desktops, to laptops, to the public Internet and nascent World Wide Web, to the modern web, to smartphones. And yet AI is also different, a step change that defies a simple depiction as such. So what changes about parenting, and what remains constant?

The post and reply

I'll quote the relevant excerpt from the post below, in which Allie K. Miller recounts a Q&A session after a recent talk.

[A woman in the audience] says her 10-year-old son used ChatGPT to write a letter to his grandma, politely requesting a brand-new MacBook. The note was so thoughtful and well-worded, grandma was moved… and promptly bought him the laptop.

Then she asked me...

“Is he a Grammy grifter? Should I be proud? Or deeply concerned?”

There were a lot of thoughtful replies, which I recommend you read. Here's mine:

LLMs are the first technology that can proxy effectively for humans in some contexts. ChatGPT is amoral at best; if you take into account the skew of its training data and the motivations of the humans building it, you might price in a bias against morality. That leaves the human in the equation as the sole guiding moral force.

Your question boils down to your son’s capacity for morality. So the way to answer your question is for you to be the guiding parent, as it is for any other dilemma with kids. Your son (and every human) has the potential to operate at all points on the spectrum when using all technologies—and especially with LLMs, because kids can substitute AI guidance for that of a concerned, moral adult human guide.

So was he grifting Grammy? Find out what his motivation was. Find out the content and context of his prompting and interactions with ChatGPT. Then guide him and teach him, and make sure he knows what’s appropriate and inappropriate, and what AI is and is not. I can think of quite a few questions I’d ask my own son, and what I’d say in response.

I guess my answer is you should be both proud and deeply concerned. And thus exactly as most parents feel at most times.

Rereading it after a few weeks, I don't see anything that I'd revise now. I'll add a few of the questions that I'd ask my son in a conversation that would include a review of his chat transcript. I'll assume that my wife and I didn't know about this project until it was a fait accompli.

Initial probing

• Did you think about whether I or Mom would enjoy being a part of the project?
• What if you had asked me or Mom for help? Would you have asked us to help in a different way? Would you have drawn different boundaries—what you wanted us to do and what you wanted to do yourself? Do you think we would have drawn our own boundaries—what we'd be willing to do and what we'd insist you do on your own?

Asking why

• Why did you decide to write a letter instead of, for example, having a conversation with her?
• Why did you think about using ChatGPT to write your letter?

Examining the what

• Who did most of the work—you or ChatGPT?
• How do you think about the output of ChatGPT? Is it your work? The AI's work? Both?

Philosophical inquiries

• If you did less work than ChatGPT, did you really earn the laptop?
• Do you think it matters whether or not Granny knows you used ChatGPT?
• What's the difference between using an AI for help and using it to do something for you? What makes it OK to use AI for one or the other? Let's come up with some examples of when it's OK and not OK for both uses.
• Is using AI different than using other technology tools, for example, using a word processor to write rather than writing by hand, or using a painting app to make art rather than using real paint and paper?

Thinking about AI

• What are some differences between chatbots and humans?
• Can AI make mistakes? If so, are its mistakes different than the kinds of mistakes humans make?
• What does it mean to trust a chatbot? What does it mean to trust a human?
• What does "doing the right thing" mean to you?
• What might "doing the right thing" mean to a chatbot?

The practicalities

My son is six. By the time he's 18, there's a strong possibility that we'll be living in an unrecognizable world—unrecognizable in the most fundamental ways. Nobody can contend with that; we just have to be ready to adapt. My current strategy with parenting, where technology is concerned, is to avoid projecting or worrying too far into the future. As some point down the road, I just get to 🤷🏻‍♂️, and then it's a completely fatuous exercise. I focus on the present and near future, and on the few constants I can identify.

At present, my wife and I do next to nothing with LLMs with our son. We both use them extensively for work. When we're curious about things and would typically refer to Wikipedia, sometimes we'll ask a chatbot; occasionally, he'll be in the mix for those kinds of conversations. But I've been thinking more about how to model proper use of LLMs with him, and we'll probably start to do that soon. He's only in first grade, so AI isn't present at school in a meaningful way.

In the next year, it's likely AI will make real leaps, and perhaps approach the realm of science fiction (read with caution if you want to sleep tonight). Leaving aside what I can't possibly predict, I know that I'll want to begin helping him get accustomed to using LLMs in a productive way. I want to do that without jeopardizing his capacity for deep thought, reasoning, creativity, and ability to anchor decisions to a set of values. I don't really have a detailed strategy, but I'll be working that out gradually (probably with the help of LLMs).

As for constants, that brings me back to my reaction to the LinkedIn post. A human world will always need people with a foundation of morals and values, and the capacity for empathy, love, curiosity, creativity, and critical thought. Those are the human traits I value the most, and as long as a society exists to practice them in, I'm going to do my best to ensure my child will thrive in it.