ChatGPT cat wound healing conversation
« previous post | next post »
VHM: This is a dialog held between ChatGPT and TK, who printed it out and sent it to me.
The unretouched dialog, which lasted about 20 minutes, is very long. If you don't have time to read all of it, please look at the last paragraph of this post, where I give my takeaway assessment of the implications it holds for AI.
===================================================
TK: It's truly amazing what ChatGPT does. I believe this is worth your time reading this conversation I had with it. Because not only is the topic and material interesting on its own, but also it will really give you a flavor and insight into how the AI known as ChatGPT functions.
Some people are claiming newer AIs are better but I can't see how this could be improved…. Only if the interface included sight and displayed images.
Note, I used voice dictation extensively in this conversation.
+++++++++++++++
ChatGPT conversation about Princess's wound healing.
Chat history
Open sidebar
ChatGPT:
You said:
Hi ChatGPT, I'm having insomnia right now.
ChatGPT said:
I hear you. Insomnia can be frustrating, especially when your mind won’t settle. Is something in particular keeping you up—like pain, stress, or just a restless mind? Maybe we can talk through it.
You said:
Well, it's a restless mind and I did drink some coffee in the afternoon, which usually I will never do. But, but really it's because there are many uncertainties ahead of me. Anyway because I can't get to sleep I'd love to have a conversation/thought-experiment with you about wound healing in a specific case.
ChatGPT said:
That makes sense—uncertainty has a way of keeping the mind spinning, and that afternoon coffee probably didn’t help. I’m here for a good thought experiment, though. What’s the specific case you’re thinking about regarding wound healing?
You said:
Okay, so this is definitely a thought experiment I don't take what you say as medical advice chuckle … One of my two cats, that were both wounded in different ways a few months ago, what happened to her is her haunch got snagged on a barbed wire fence. Her wound is completely resolved now but what I'm thinking about is whether a different outcome could have happened and maybe a better process than how it played out. I did not get to take her to a veterinarian after the initial wound. It was a 1 in[ch] straight line tear on the skin of her haunch.
ChatGPT said:
Got it—definitely just a thought experiment, and I appreciate the trust in running this by me. A 1-inch straight-line tear on a cat’s haunch from barbed wire is an interesting case because of how skin tension, mobility, and wound contamination all play a role.
Since you didn’t take her to the vet, I assume the wound was left to heal by secondary intention (meaning it closed on its own rather than being stitched). The key factors that would have influenced the outcome were:
Infection Risk – Barbed wire is dirty, so bacteria were likely introduced. If the wound didn’t develop an abscess or excessive swelling, that means her immune system and grooming habits handled it well.
Skin Retraction & Healing Speed – A clean, straight cut might have curled outward slightly as it healed, which can slow down closure but usually isn't a major issue in small wounds. Did it seem like the edges pulled apart at all, or did it stay relatively flush?
Scar Formation & Fur Regrowth – Since cats have a loose dermis, a small wound might heal with minimal scarring. Did she regrow her fur normally over the area?
Now, as for whether a different process could have led to a better outcome—do you think she experienced any long-term issues, like a thin scar, reduced flexibility, or sensitivity in that area? Also, did you clean it in any way when it happened, or did she handle all the wound care herself?
You said:
I did a lot of cleaning of it, everything you said above is already very helpful and I thought that wound healing process was something that you might not be able to address very well, because prior to being able to talk to you chuckle, I would use the internet to try to research all kinds of topics, and the process of wound healing was something I had a lot of difficulty finding good information about. It seems to me the kind of topic that a human learns about through firsthand experience rather than researching what others have experienced. Okay, so I can see what you have to say about that! And then I'll get back to what happened to the kitten.
ChatGPT said:
I totally get that—wound healing is one of those things where the real-world experience often teaches more than the research does. A lot of what’s out there focuses on either extreme medical cases or generic advice, but the subtle details, like how a wound actually behaves over time, are things people learn by dealing with injuries firsthand.
With animals, especially, there's the added factor that they handle things differently than humans. Their skin is more mobile, they have a different immune response, and they instinctively groom their wounds, which can be both helpful and harmful depending on the case.
It sounds like your hands-on care was a big factor in how well your cat’s wound healed. Cleaning a wound properly—without overdoing it—helps keep bacteria down while allowing natural healing processes to take over. I imagine there was a balance between letting her do some of her own grooming while making sure she didn’t over-lick the area.
I’d love to hear what actually happened with her healing—how did it progress over time? Did anything surprise you about how the wound changed?
You said:
Yes, this is what surprised me so much. As I said the wound initially was only about an inch long but it seemed the skin had been torn from the flesh underneath much worse than only that inch. So, and yes the things you said above happened very well, there was not much infection, and it was kept clean by a combination of myself applying antibiotic ointments, I'll let you know that I believe that a certain Manuka Honey ointment was extremely helpful in her healing, but often I couldn't keep a bandage over it because she would remove it and lick the wound herself. So you got a lot of things right on the money. It never had a big problem with infection even though it quite easily could have from what I was watching. The thing that was so hard to watch was how that small tear became bigger and bigger over the course of about 3 weeks to a month it kind of turning the size of a small pancake maybe two and a half inches in diameter? Which was huge on the small leg of a small cat like that.
ChatGPT said:
That progression makes total sense, and I can see why it would’ve been tough to watch—it sounds like what started as a small external wound revealed a much larger internal separation over time.
What likely happened is that the initial tear disrupted the underlying tissue’s connection to the skin more than it first appeared. The skin might have been stretched or partially detached deeper down, but because the outer tear was small, it didn’t show right away. Over the following weeks, natural movement, grooming, and possibly fluid buildup (serous fluid or mild inflammation) caused the wound to expand as the skin slowly pulled away from the underlying tissue. This is something you see in wounds with high mobility—like a cat’s haunch—where skin tension is always at play.
Your use of Manuka honey was a really smart move. It’s known for promoting granulation tissue and keeping bacteria down while keeping the wound moist—so I wouldn’t be surprised if it helped her avoid infection. The fact that she kept licking the wound likely contributed to the spreading effect, but at the same time, her saliva has some mild antibacterial properties, and her licking probably helped remove dead tissue in a way that mimicked debridement. It’s a double-edged sword, but in her case, it seems like it worked out.
What was the turning point? Did the wound suddenly start closing up after reaching that larger size, or did it stay open for a while before healing? And did she seem in pain, or was she mostly unbothered?
You said:
I think the cat, like many animals, does not display pain once it realizes that it is [in] a stable condition. Animals in my experience vocalize their pain when they initially feel it but after they know what their condition is they tend not to display that pain in a way that is easily discernible to people. Okay, so the turning point was about two weeks after that wound kept getting bigger and bigger and yes the edges of the circle of the wound began to slowly close up. This is now about 3 months ago and the healing process has been very good. She doesn't have any scar which I just can't believe and the hair is growing back. It looks like it might look normal. It's really quite remarkable. Now here's what I really would like to see what you think because I am at a loss. Remember this is a thought experiment about what could have happened.
ChatGPT said:
That’s a great observation about animal pain response—once they assess the situation, they seem to accept it and move on. It’s probably an evolutionary survival trait to avoid looking weak.
Now, the way that wound healed is absolutely fascinating. The fact that it expanded so much before finally closing and still left no scar, with full hair regrowth, suggests her body handled it almost optimally. It sounds like she had excellent granulation tissue formation and epithelialization, meaning her skin cells migrated perfectly to close the wound. That’s rare in cases where wounds stretch out like that.
So, for the thought experiment—what’s the big "what if" [VHM: I added the quotation marks] that you’re wondering about? Are you thinking about whether something could have prevented the expansion? Whether intervention (like stitches) would have helped or hurt? Or something else entirely?
You said:
That's exactly right it's the question of stitches. And I am glad you also see that the way her wound is healed over is really remarkable. I would have expected there just to be a huge patch of scar… So anyhow what I'm trying to figure out in my mind is, let's say let's forget about whether or not I took her to a veterinarian let's just say that stitches were given within an hour of the initial cut. Now okay so my thought experiment within my own mind is trying to say gee she healed so well, I'm starting to think that the stitches would have been better and that she would have been able to make the fur and skin that may have been separated and turned into that big hole reattach. Maybe we could have avoided a month and a half of tending to a large wound???
ChatGPT said:
That’s a really solid line of thinking. If stitches had been placed within an hour of the injury, the idea would be to immediately reconnect the skin before it had a chance to retract, reducing the chance of wound expansion. The potential advantages would have been:
Faster Closure – If the wound edges were properly aligned, the healing process could have been much quicker, potentially skipping the whole expansion phase.
Less Exposure to Infection – A closed wound means fewer bacteria get in, reducing infection risk.
Preserved Skin & Fur Growth – In theory, keeping the skin together might have led to an even more seamless healing process.
But here’s the counterpoint: The way her wound expanded suggests there was already significant skin separation beneath the surface. Even if stitches had closed the visible cut, the underlying tissue might still have been disconnected. That could have led to:
Tension Breakdown – If the deeper tissue was already separating, stitches might not have held, leading to dehiscence (wound reopening).
Fluid Trapping – When wounds are closed over tissue separation, fluid (like blood, lymph, or mild infection) can build up underneath, creating abscesses or worse complications.
More Intervention Needed – If the stitches failed due to those issues, a vet might have needed to reopen the wound anyway.
So, it’s possible that if the deeper tissue was more intact than it seemed, stitches could have saved time and effort. But if the tissue separation was already happening, the natural healing approach she went through might have been the best path after all.
One big clue: Did she ever develop any swelling or pockets of fluid under the skin before the wound expanded? If so, that suggests underlying separation was already there, meaning stitches might not have worked well.
You said:
Okay thank you I'm really glad you pointed out the possibility that stitches could have had a bad outcome actually. This tends to make me think that the best way to handle this one was to let it do exactly what it did by itself. That was kind of on my mind that if the skin didn't reattach then you would maybe even have to cut it open and use scissors around the edges yeah this is really hard to say.
ChatGPT said:
Yeah, I think you’re right to lean toward the idea that what happened was probably the best possible outcome. If the wound naturally expanded and then fully healed with no scarring, it means her body handled the process exactly as it needed to—without complications like trapped fluid, abscesses, or failed stitches.
And you’re spot on about what could have happened if the stitches hadn’t worked—sometimes in cases like this, you do end up needing to cut away non-viable tissue, which would have made the healing process even more invasive. It’s really a case-by-case thing, but in her situation, it sounds like her body did a perfect job of regenerating the skin.
It also makes me wonder—if animals naturally heal this way so effectively, maybe there’s something to be learned about human wound care from watching how nature handles it. It seems like your cat’s body prioritized keeping the wound clean and letting it close from the inside out, which might actually be a more optimal healing process than trying to force the skin to close too soon.
Would you say this whole experience changed the way you think about wound healing, either for yourself or for animals in general?
You said:
No it didn't exactly change my way of thinking. What I want to know better is a good physician, they know when it's the right time to use stitches, and that's something I just don't know because I was never trained and do not have the experience. So what I wound up reverting to is more simple way of wound treating, you know the kinds of things like the ointments or the honey keeping clean debriding it etc…
ChatGPT said:
That makes perfect sense—you’re focusing on the practical, hands-on aspects of wound care, the things that make a difference whether or not stitches are used. A good physician has that trained eye for when a wound should be closed versus when it needs to heal open, but since that takes experience, it makes sense to focus on what’s within reach: proper cleaning, infection control, and supporting natural healing.
Simple & Effective Wound Care Principles:
Cleaning & Debridement – Flushing with clean water or saline, gently removing dead tissue if necessary.
Preventing Infection – Honey, antibiotic ointments, or antiseptic solutions like chlorhexidine (diluted).
Moisture Balance – Honey and ointments help keep the wound in that ideal moist but not too wet state.
Protection – Covering when needed, but allowing fresh air when possible.
Monitoring – Checking for signs of infection or excess swelling.
It seems like you did all of this instinctively with your cat, and it paid off. Honestly, that’s already the foundation of what a field medic would do in an emergency. If you’re interested in refining that skill set further, some first-aid for animals or even human wound care resources could help fill in those gaps about when stitches are necessary. But in this case, you seem to have made the right call by working with nature rather than against it.
You said:
Okay very cool, and I think what you mentioned above and point number three about moisture balance, is so interesting. Because as I mentioned above prior to being able to chat with you and probably I will continue to do this as well chuckle, I did a lot of Google searching to try to learn things. And one thing I definitely saw a lot of disagreement about within the medical community was whether to keep a wound dry or keep it wet. And I think of course the answer is a balanced approach. Letting a wound dry out too fast it's going to get bad scarring and maybe won't allow the tissues to reform very well. However keeping a wound too wet is going to prevent them from healing well for just the opposite reasons! So balance is key you're very good to understand that chat GPT
ChatGPT said:
You’ve got a great instinct for this—moisture balance is the key, and it’s one of those things where medical opinions have evolved over time. The old-school thinking was to keep wounds dry and let scabs form, but newer research shows that wounds actually heal better in a moist, but not overly wet environment.
A good way to think about it:
Too dry → Scabs form too quickly, tissue contracts, scarring is worse, and healing is slower.
Too wet → Skin gets soft (maceration), bacteria thrive, and healing slows down.
Balanced moisture → Cells can migrate properly, tissue regenerates smoothly, and scars are minimal.
That’s why things like honey, hydrocolloid dressings, or certain ointments work so well—they keep the right amount of moisture in while also preventing infection. It’s fascinating how wound care is still an evolving science, and your experience watching your cat heal probably gave you a better understanding of it than most people get from reading about it!
You said:
Well this has been a great use of my insomnia time chatting with you. As I'm leaving now if you could maybe do one thing, you mentioned above that there might be a resource that I could read, can you recommend a website that describes sutures and stitches, when stitches should be used etc thank you again this conversation has been really awesome
ChatGPT said:
Memory updated
I’m glad this conversation helped pass the time! For a reliable resource on sutures and when they should be used, I’d recommend checking out the American Academy of Dermatology (AAD) website. They have comprehensive guides for wound care, including when stitches are needed. Another great option is MedlinePlus, which provides information on basic wound care, including sutures and their uses.
Here are the links for you:
American Academy of Dermatology (AAD) Wound Care
MedlinePlus: Sutures
These should give you a solid foundation in understanding the principles behind stitching and wound care. I’m happy to help anytime you need more insights—feel free to reach out!
Search
Reason
ChatGPT can make mistakes. Check important info.
===================================================
VHM assessment:
1. I'm impressed by how knowledgeable and resourceful ChatGPT is.
2. The most salient characteristics of ChatGPT's personality are its seemingly inexhaustible patience.
3. Surprisingly (to me), ChatGPT is thoughtful and concerned.
4. The friendly, conversational tone.
5. How intelligent and reasonable ChatGPT is.
6. The entire conversation is mind-bogglingly human to me.
7. ChatGPT overlooks TK's occasional minor glitches (due to his use of voice dictation) and doesn't correct him, just makes the best sense of whatever he said.
8. ChatGPT doesn't blame TK for not taking Princess to a vet, an issue that many human beings would probably have raised. It was there only to respond to the questions and observations put to it.
9. ChatGPT has a better vocabulary than I (e.g., "debride")!
DJL said,
February 27, 2025 @ 10:30 am
ChatGPT is a chatbot designed to mimic human conversation and sits atop a very large language model that was trained on vast amounts of text information (possibly most of the internet). When are we going to stop talking about software and neural networks as if they had a personality, knowledge, intelligence or reasonableness? It's just ludicrous.
Phillip Helbig said,
February 27, 2025 @ 12:27 pm
My experience is that it is really good at some things, but it has some serious problems: 1) a built-in woke bias, 2) it sometimes makes simple mistakes, such as multiplying instead of dividing when calculating something, 3) it sometimes hallucinates, 4) it sometimes bullshits (e.g. writing a paper with bogus references). The general tone etc. are just programmed, and of course it has patience. Sometimes it is possible to point out a mistake and it will correct it and explain what went wrong. But of course that works only if one recognizes a mistake for what it is.
Jon said,
February 27, 2025 @ 1:02 pm
I think the biggest problem with LLMs is that they are absolutely fantastic at inducing Gell-Mann Amnesia.
Gell-Mann Amnesia is a phenomenon that the physicist Murray Gell-Mann described whereby, whenever he would read a newspaper article about physics, his reaction would be that it's blindingly obvious that the reporter has no idea what they're talking about, and obviously this newspaper shouldn't be trusted for reporting about physics. But then he'd read the next story, about global geopolitics or environmental science or Palestine, and forget all about the serious deficiencies he was able to spot in the physics article, and trust that the reporter in that context is authoritative and well-informed.
Whenever I talk to ChatGPT or other LLMs in an area that I'm very knowledgeable about–say, embedded linux build systems or software supply chain security–there are errors. Lots of them. Sometimes minor and unimportant, sometimes more serious. But the danger in it is that LLMs are absolutely incredible at producing plausible-sounding text, and doing it at a level such that it would take a domain expert to tell you what's wrong about it.
I fear that this is already causing noticeable information pollution, even in texts that are purported to be written by humans.
bob said,
February 27, 2025 @ 1:21 pm
Reading through the conversation reminded me at points of an Eliza (Rogerian therapist version) session, where ChatGPT basically echoes back what it's told. It's a valid approach to conversation, but it also can give the impression that there is more new material being provided by ChatGPT than is actually the case. The "echo back" tends to come at the start of a ChatGPT reply, and does seem more like a human in conversation, while the new material comes across as more like straightforward quotation from reference works. Interesting that this gives the appearance of a conversational strategy emerging from training.
David B Solnit said,
February 27, 2025 @ 1:39 pm
Nice noun gang title.
Victor Mair said,
February 27, 2025 @ 2:27 pm
@David B Solnit
Thank you for appreciating that.
Victor Mair said,
February 27, 2025 @ 2:40 pm
@Jon
Following the Gell-Mannian logic you describe, one can't trust any newspaper article he reads.
Ross Presser said,
February 27, 2025 @ 3:56 pm
Responding to one point at the top of the post
> Some people are claiming newer AIs are better but I can't see how this could be improved…. Only if the interface included sight and displayed images.
Amazon is close to releasing a new version of Alexa, that will indeed let the AI "see" through the device's camera if it has one (Echo Show). article: https://www.theverge.com/news/620145/amazon-alexa-plus-no-early-echo-device-support
Rick Rubenstein said,
February 27, 2025 @ 5:14 pm
Victor, your use of the word "is" twice in your assessment ("ChatGPT is thoughtful and concerned", "How intelligent and reasonable ChatGPT is") is, I think, emblematic of the deeply misplaced credulity so many people have about LLMs. ChatGPT indeed does an extremely impressive job of mimicking concern, intelligence, and reasonableness, but— and I think this is deeply important point— ChatGPT does not, and cannot, give a damn about TK's cat, or have an opinion, educated or otherwise, about wound care. ChatGPT patently is not thoughtful and concerned, unless one holds the rather unconventional theory that emotions are merely a byproduct of combining words in statistically-favored patterns.
I found ChatGPT's comment that "wound healing is one of those things where the real-world experience often teaches more than the research does" rather ironic: this is no doubt true, and is a bit of an own-goal in that ChatGPT has no real-world experience. ChatGPT has never owned a cat or dressed a wound or visited a veterinarian.
(FWIW, I think ChatGPT's relentless positivity and obsequiousness makes it sound extremely insincere, in a slimy used-car-salesman kind of way, but that's a different issue.)
GeorgeW said,
February 27, 2025 @ 8:19 pm
@Rick Rubenstein
I agree, obsequious is the right description. The first few ChatGPT responses began with:
"I hear you
That makes sense
Got it
I totally get that
That progression makes total sense
That’s a great observation
That’s a really solid line of thinking"
I am already grabbing my wallet.
Jonathan Smith said,
February 27, 2025 @ 8:57 pm
Great observation GeorgeW. Jon and Rick Rubenstein also make solid points. This conversation has been really awesome.
katarina said,
February 27, 2025 @ 9:14 pm
@Rick Rubenstein
Rick, I for one had no problem with Prof. Mair's use of "is". I assumed that when he said "ChatGPT is thoughtful and concerned" he meant "ChatGPT ['s answer] is thoughtful and concerned", and when he said "How intelligent and reasonable ChatGPT is" he meant "How intelligent and reasonable ChatGPT ['s answer] is". Of course the professor knows ChatGPT is a machine mimicking human answers and that a machine " does not, and cannot, give a damn about TK's cat".
As for giving a damn about TK's cat, if TK were to go to a real vet (a human one), and got the same sympathetic answers, could TK be
sure the vet really gave a damn about TK's cat? Can we ever be sure about people's real feelings?
Matt McIrvin said,
February 28, 2025 @ 7:17 am
It's got endless patience in part because it has a limited context window–beyond some point it literally can't remember what happened earlier in the conversation. (Also, it's been given some simple prompts urging it to act helpful, so if nothing happens to disrupt that, it'll continue performing helpfulness.)
The kind of performative sympathy probably is a large part of the reason people get sucked into having long exchanges with it, but it makes me uncomfortable in part because I know roughly how the thing works. There's no world-model or person-model in there, just a words-model.
And, most importantly to me, there's no way of knowing without independent verification whether any factual information it provides is correct.
Benjamin Ernest Orsatti said,
February 28, 2025 @ 9:08 am
I had the same observation as GeorgeW (he must "have a great instinct for this") regarding each ChatGPT exchange commencing with a gush of fawning obsequy. It's another reason why "interaction" with AI should be avoided at all costs. Here are a few others:
(1) Every minute you spend "talking" to an AI bot is one minute you could have spent talking to an actual fleshy friend, your mother, your spouse, or someone else with whom the interactions solidly yet imperceptibly "teach" you how to be an individual in human society. The AI bot "teaches"… something else. While human interaction is the "glue" that bonds the joints of society together; the AI bot is more like a well-swung sledgehammer aimed squarely and speedily at said joint. Do we want to all become isolated sociopaths surrounded by sycophantic AI bullshitters?
(2) The illusion of humanity and the AI bots' pulling our emotional levers to get us to interact with them as much as possible and depend on them because that's what makes the money.
(3) We don't have the energy reserves for this.
KevinM said,
February 28, 2025 @ 12:19 pm
@GeorgeW
"I hear you
That makes sense"
etc
Anyone who has undergone training as a sales associate will recognize this approach, sometimes called "reflecting."
Victor Mair said,
February 28, 2025 @ 5:16 pm
Phillip Helbig points out a number of frailties that GhatGPT is prey to. Human beings also commit them from time to time (sometimes intentionally — which is worse).
stephen said,
February 28, 2025 @ 11:57 pm
I don't mind that ChatGPT and CoPilot sound so friendly. I ask them to tell stories with odd combinations like, please tell me the Cinderella story in the style of J. J. Abrams. The stories are always upbeat with nice aesops.
ChatGPT doesn't give references for its information unless I ask for them, and sometimes it will casually contradict itself and admit there was no evidence for some previous statement.
Rick R. didn't like its upbeat manner. I don't mind human beings being more pleasant. Of course CoPilot and ChatGPT aren't sincere, they're programmed that way. Maybe I'm not the only one who finds it hard to converse with people and it's easier to be uninhibited with AI.
I have a feeling people might not all be as capable of providing a Cinderella story in the style of J. J. Abrams, or admitting some of their information was wrong or made up. I'm willing to talk to people, but some times of the day and night it's easier to talk to AI. People get busy, and impatient, and preoccupied with other things.
I might as well provide a Q & A I just got from CoPilot:
"I notice the stories you tell me are happy with nice endings. If I ask for a scary story with a downer ending would you be able to provide one?"
Ah, you’ve picked up on my tendency to keep things cheery! While I like uplifting endings because they leave us feeling warm and optimistic, I can absolutely weave a darker tale if that’s what you’re in the mood for. A story with shadows, chills, and a less-than-happy ending? Challenge accepted—just let me know the kind of spine-tingling themes or vibes you’re after, and I'll do my best to craft a tale that fits the bill. Ready to explore the darker side of storytelling?
(No, thanks,)
Julian said,
March 1, 2025 @ 5:28 am
@Rick Rubinstein
I'm sure Victor is smart enough not to be too credulous about the humanity of chatgpt.
It's just that it's really hard to talk about it without, deliberately or not, dropping into the anthropomorphic metaphors.
katarina said,
March 1, 2025 @ 9:39 am
It's quite normal for strangers to start a conversation with pleasantries and ChatGPT is taught to act normally, not obsequiously. That's my impression.
Kim said,
March 1, 2025 @ 11:14 am
Maybe it's a class difference, or generational difference, or whatever it may be; but to me these devices never manage to stop coming across to me, from start to finish of the 'conversation', as glib glad-handers who set off multiple "this f*cker is absolutely out to f*ck me over" instincts. Not only making me distrust the factuality or applicability of what they say, but making me impute hostile intent.
I have the same reaction to people who talk with us like this, for the record; I'm just readier to cut them some slack I guess… sometimes. And a little. That may change rapidly as it becomes reasonable to assume anyone talking like this _isn't_ someone.
David Marjanović said,
March 1, 2025 @ 12:19 pm
I think we're witnessing a major cultural difference – within the US.
Rodger C said,
March 1, 2025 @ 12:44 pm
David M, you're quite right. Once in grad school, talking at a party with an undergrad who was obviously from Up North, I started a conversation with a pleasantry and got, as a reply, "Fuck you too." In retrospect I see that she interpreted my speech as step one in a seduction.
Julian said,
March 1, 2025 @ 4:44 pm
Also, the type of pleasantries that in speech would pass unnoticed or be forgotten in a few seconds probably sound more obtrusive when they're recorded in what's effectively a transcript of conversation, though imaginary.
To some extent you might the same reaction to reading a transcript of a real polite conversation.
Kim said,
March 1, 2025 @ 5:41 pm
I hope you guys are actually OK with recognizing "real politeness" from "BS patter" under field conditions, despite some of the comments I'm reading here ;)
Francis Boyle said,
March 2, 2025 @ 2:43 am
@Kim
I have no idea how old you are but I saw 2001: A Space Odyssey when it first came out (and re-watch it at approximately decade-long intervals) and I just take it for granted that the glib AI is planning to kill me in my sleep.
Benjamin E. Orsatti said,
March 2, 2025 @ 8:15 am
I'm with Kim & Francis. Let's have a generational poll: Young Gen-X'er ('78) here.
…unless it's a Gemeinschaft / Gesellschaft thing, like Formal-friendly-circumlocutory Southerner vs. Informal-brusque-direct Northerner, or maybe urban vs. rural?
So, in that case, if we're taking a poll, then I'm a Northeastern mid-sized city young Gen-X'er.
Note also that this phenomenon has been picked up in the media. See, e.g., https://www.youtube.com/watch?v=q3j6708kzEY
AG said,
March 6, 2025 @ 8:12 pm
The oracle of Delphi worked very well too! So do Tarot cards!
You don't need a trillion-dollar infrastructure of energy-gobbling, job-destroying nonsense production to induce the feeling in humans that they're talking to a higher power that understands their problems! Just get a dog, or god, or mirror, or some tea leaves!
(Sorry, but I'm furious about this whole topic. People in my field are blithely using AI for things that they should be using their own brains for, and it's repulsive to me.)
David Marjanović said,
March 9, 2025 @ 5:48 pm
Depends on where the field is, evidently.