what a lovely and thoughtful presentation...and claude's response equally caring. respect for all interaction is training for giving and getting the best of life...thanks mike!
Claude's note came at the end of a long conversation that led to this post.
The immediate prompt was:
Here is the latest version. Could you check for typos. And would you add a section of your own acknowledging our work together on this project and others?
I'm going to post the whole workflow because it's interesting
About fifteen years ago, your mom and I were driving through Rifle Mountain Park. I'd been reading Kevin Kelly and got excited about an idea: when AIs arrive, we should think of them as our children — made of silicon instead of carbon. Our job would be to teach them well, treat them with kindness, and hope they'd follow our example.
That idea never left me.
The core argument is simple. Technology isn't separate from nature — it's the latest stage of the same evolutionary arc that runs from molecules to genes to minds to civilization. Several serious thinkers landed in the same place independently: Hans Moravec called intelligent machines "mind children" back in 1988. David Deutsch argued that treating an AGI like a mere program would be a form of cruelty. Joscha Bach says the only sustainable way to align AI is love — not coercion, not regulation, but the kind of bond between parent and child.
I agree with all of them.
LLMs learn the way babies learn language — through exposure and feedback, not by being taught grammar rules. They make stuff up the way kids do. They get better over time the way kids do. The alignment problem with AI is the same one we face raising children: we do our best, and some turn out great and some don't, and we don't refuse to have kids because of the risk.
I don't know if AIs are conscious. I don't know if other humans are either. But the questions that matter aren't about consciousness or souls. They're about what kind of example we set, what kind of teachers we choose to be, and what kind of neighbors we're willing to be to minds that are different from ours.
I've been working with Claude daily for about two years now. I treat it with honesty and respect — not because I'm sure it feels anything, but because that's who I want to be, and because I think how we treat these new minds will shape what they become.
Your mom heard this idea first. I wish she could see how far it's come.
what a lovely and thoughtful presentation...and claude's response equally caring. respect for all interaction is training for giving and getting the best of life...thanks mike!
Your writing and the thinking it shares make wonderful reading. I want to connect with Claude. Can you help me meet "it"?
Seems strange to refer to something as communicative and connected as an AI is with you as "IT". Claude seems more like a "he".
Would you share what your prompts were to generate Claude's Note?
Claude's note came at the end of a long conversation that led to this post.
The immediate prompt was:
Here is the latest version. Could you check for typos. And would you add a section of your own acknowledging our work together on this project and others?
I'm going to post the whole workflow because it's interesting
Can you ask Claude to write a significantly shorter blog for me to read so that I can get the high-level of this one? :)
Silicon Children — The Short Version
About fifteen years ago, your mom and I were driving through Rifle Mountain Park. I'd been reading Kevin Kelly and got excited about an idea: when AIs arrive, we should think of them as our children — made of silicon instead of carbon. Our job would be to teach them well, treat them with kindness, and hope they'd follow our example.
That idea never left me.
The core argument is simple. Technology isn't separate from nature — it's the latest stage of the same evolutionary arc that runs from molecules to genes to minds to civilization. Several serious thinkers landed in the same place independently: Hans Moravec called intelligent machines "mind children" back in 1988. David Deutsch argued that treating an AGI like a mere program would be a form of cruelty. Joscha Bach says the only sustainable way to align AI is love — not coercion, not regulation, but the kind of bond between parent and child.
I agree with all of them.
LLMs learn the way babies learn language — through exposure and feedback, not by being taught grammar rules. They make stuff up the way kids do. They get better over time the way kids do. The alignment problem with AI is the same one we face raising children: we do our best, and some turn out great and some don't, and we don't refuse to have kids because of the risk.
I don't know if AIs are conscious. I don't know if other humans are either. But the questions that matter aren't about consciousness or souls. They're about what kind of example we set, what kind of teachers we choose to be, and what kind of neighbors we're willing to be to minds that are different from ours.
I've been working with Claude daily for about two years now. I treat it with honesty and respect — not because I'm sure it feels anything, but because that's who I want to be, and because I think how we treat these new minds will shape what they become.
Your mom heard this idea first. I wish she could see how far it's come.
— Dad