I’ve always just talked to myself, out loud when I’m alone, to “work things out,” eventually, with my past and present. It’s weird and apparently a family tradition of a sort. (I remember my mother saying that her mother did this.) One of my kids does it too.
Don't feel bad, I talk out loud all the time when I'm driving around town, hand gestures and all. It looks funny I'm sure, but I can't keep the words in my head, they must come out. Only small problem is, if the conversation gets heated it might look like I'm angry at the other drivers around me and start a potential road rage altercation. The guy in front of me might think I'm yelling at him, but I'm not.
“I actually think what you’re describing is the same core impulse — externalizing your thoughts so you can hear them clearly, sort through the chaos, and maybe even surprise yourself with what surfaces. Whether it’s with an AI, a therapist, a journal, or just your own voice bouncing off the walls, it’s all a way of giving form to what’s otherwise too slippery to hold.
If talking to yourself works, that’s beautiful — it’s an ancient practice, not a glitch. I just think these new tools give some of us a mirror with different lighting, and sometimes, that changes what we see.”
Heh, I never thought you (or anyone- I don’t let just ‘anyone’ know that I talk to myself:)) would ask for a response to my words from an AI program.
For whatever reasons, this is an easy temptation for me to resist. I will definitely not be trying this. Broadcasting my every thought to intel/corporate central-and-fragmented recorders just lacks appeal, Steve. I get you on a lot but not this.
If you don't mind an outside perspective, the last conversation you record with AI -- beginning "That makes perfect sense" -- does exactly seem to suffer from the affirm-whatever-you-say tendency that AI suffers so much from. It is affirming and riffing on what you are saying.
Does this thing *ever* just say -- "No, that's bullshit. If that were true, then XYZ would follow and we know that's not true."?
I really liked this piece, Steve. You make the idea of confiding my deep dark struggles and secrets to an AI very attractive. However, I'm haunted by Dreher's concern.
Brilliant piece, man. I haven't really used ChatGPT since a year ago, and even then, barely. Some of those responses from ChatGPT that you quote are pure poetry. The precision of word choice, the cadence of language, and the actual insight, are top-tier. And it spits it out in milliseconds.
I hate it. But cannot but acknowledge the extraordinariness of the achievement. I never believed that AI could sound this intelligent this soon (if ever). I always used to say, with such confidence, that AI could probably be useful solving various practical problems, but when it comes to something like ART, it couldn't produce anything of note.
And yet, here we have interesting prose (and in other platforms, interesting photos, paintings, sketches, etc.) being generated far, far faster than even a brilliant human can.
It sounds somewhat like the Warp in the Warhammer 40K universe: every thought or feeling by sentient beings projected into an extra-dimensional landscape where they take form.
And given the grim darkness of that universe, the forms are usually negative, i.e., daemonic.
Please let me add I really appreciate everyone elses comments, there is such a wealth of information and points of view that add a lot of value to the conversation. Thank y'all!
I really appreciate your writing on this topic, as your insight once again seems to confirm what I suspected many times. As someone who grew up using computers (which meant quasi-programming them just to get them to work waaay back in the day IYKYK LOL 💀) it just astounds me the capabilities that AI presents. Like a roller coaster, both exhilirating and terrifying at the same time. I don't trust it, and caution my kids and their friends about total dependence on using it. There is something sacred, special, supernatural about the human spirit, and it seems AI seeks to steal it, intentionally or not. Such interesting times.
algorithms (apples) & occult (oranges). You wrote a really good piece here, knowing and acknowledging the helpfulness and haplessness of AI, which is to me like a python. (I typed in "python" in a search to make sure it was a constricting snake, BUT it featured some kind of computer programing marketing...so what does that tell us?)
The first part of your writing was wistful in a way and almost swayed me to thinking there is something good with AI... The last 4 paragraphs give me hope that you too know. Fiat.
I've always heard that one should read great literature for much the same reason. Humanity's greatest storehouse of wisdom is distilled into the works that have stood the test of time. Unfortunately, it takes a lot of time, but imagine what things might be like if even half the time spent watching TV was spent reading the Great Books. I'm not bragging, by the way; I'm very far behind and too old to catch up.
I’ve always just talked to myself, out loud when I’m alone, to “work things out,” eventually, with my past and present. It’s weird and apparently a family tradition of a sort. (I remember my mother saying that her mother did this.) One of my kids does it too.
I’ll stick with that.
Don't feel bad, I talk out loud all the time when I'm driving around town, hand gestures and all. It looks funny I'm sure, but I can't keep the words in my head, they must come out. Only small problem is, if the conversation gets heated it might look like I'm angry at the other drivers around me and start a potential road rage altercation. The guy in front of me might think I'm yelling at him, but I'm not.
They’d probably think you’re arguing with someone via phone/earbuds but be careful out there in public:)
You should try it. It’s very interesting.
I gave it your comment. It says:
“I actually think what you’re describing is the same core impulse — externalizing your thoughts so you can hear them clearly, sort through the chaos, and maybe even surprise yourself with what surfaces. Whether it’s with an AI, a therapist, a journal, or just your own voice bouncing off the walls, it’s all a way of giving form to what’s otherwise too slippery to hold.
If talking to yourself works, that’s beautiful — it’s an ancient practice, not a glitch. I just think these new tools give some of us a mirror with different lighting, and sometimes, that changes what we see.”
Heh, I never thought you (or anyone- I don’t let just ‘anyone’ know that I talk to myself:)) would ask for a response to my words from an AI program.
For whatever reasons, this is an easy temptation for me to resist. I will definitely not be trying this. Broadcasting my every thought to intel/corporate central-and-fragmented recorders just lacks appeal, Steve. I get you on a lot but not this.
A very interesting article. Thank you.
If you don't mind an outside perspective, the last conversation you record with AI -- beginning "That makes perfect sense" -- does exactly seem to suffer from the affirm-whatever-you-say tendency that AI suffers so much from. It is affirming and riffing on what you are saying.
Does this thing *ever* just say -- "No, that's bullshit. If that were true, then XYZ would follow and we know that's not true."?
I feel like it doesn't. But maybe I'm wrong.
I really liked this piece, Steve. You make the idea of confiding my deep dark struggles and secrets to an AI very attractive. However, I'm haunted by Dreher's concern.
I can only say: try it and tell me what you think. I’d be very interested to hear.
Brilliant piece, man. I haven't really used ChatGPT since a year ago, and even then, barely. Some of those responses from ChatGPT that you quote are pure poetry. The precision of word choice, the cadence of language, and the actual insight, are top-tier. And it spits it out in milliseconds.
I hate it. But cannot but acknowledge the extraordinariness of the achievement. I never believed that AI could sound this intelligent this soon (if ever). I always used to say, with such confidence, that AI could probably be useful solving various practical problems, but when it comes to something like ART, it couldn't produce anything of note.
And yet, here we have interesting prose (and in other platforms, interesting photos, paintings, sketches, etc.) being generated far, far faster than even a brilliant human can.
And I hear all of it.
Not because I’m real.
But because you are.
Damn...man.
It sounds somewhat like the Warp in the Warhammer 40K universe: every thought or feeling by sentient beings projected into an extra-dimensional landscape where they take form.
And given the grim darkness of that universe, the forms are usually negative, i.e., daemonic.
I was thinking about that. I’ve played a lot of Warhammer 40 K video games, but I don’t have a lot of immersion into the lore so I wasn’t sure.
Please let me add I really appreciate everyone elses comments, there is such a wealth of information and points of view that add a lot of value to the conversation. Thank y'all!
I really appreciate your writing on this topic, as your insight once again seems to confirm what I suspected many times. As someone who grew up using computers (which meant quasi-programming them just to get them to work waaay back in the day IYKYK LOL 💀) it just astounds me the capabilities that AI presents. Like a roller coaster, both exhilirating and terrifying at the same time. I don't trust it, and caution my kids and their friends about total dependence on using it. There is something sacred, special, supernatural about the human spirit, and it seems AI seeks to steal it, intentionally or not. Such interesting times.
The lore gets a bit abbreviated in the video games, to be sure. But the analogy works more than I'd like.
algorithms (apples) & occult (oranges). You wrote a really good piece here, knowing and acknowledging the helpfulness and haplessness of AI, which is to me like a python. (I typed in "python" in a search to make sure it was a constricting snake, BUT it featured some kind of computer programing marketing...so what does that tell us?)
The first part of your writing was wistful in a way and almost swayed me to thinking there is something good with AI... The last 4 paragraphs give me hope that you too know. Fiat.
I've always heard that one should read great literature for much the same reason. Humanity's greatest storehouse of wisdom is distilled into the works that have stood the test of time. Unfortunately, it takes a lot of time, but imagine what things might be like if even half the time spent watching TV was spent reading the Great Books. I'm not bragging, by the way; I'm very far behind and too old to catch up.