[HN Gopher] You sent the message, but did you write it?
___________________________________________________________________
You sent the message, but did you write it?
Author : ano-ther
Score : 48 points
Date : 2025-05-03 20:08 UTC (2 hours ago)
(HTM) web link (davidduncan.substack.com)
(TXT) w3m dump (davidduncan.substack.com)
| turtleyacht wrote:
| Hmm. Almost makes sense to write emails
| with deliberate newlines--not counting the widths, but
| advancing the line as it were (ourselves), like how we'd do
| with a typewriter. Because (in Outlook) I didn't like
| seeing the message stretch out past where my eyes could
| scroll too far right; instead, to Enter when each
| line sort of looked alright, some length among their
| sentence homes.
| nkrisc wrote:
| Oh, come on, please, no. You have no idea on how wide of a
| window or screen I'm reading it with or what my font size is.
|
| Just write text, separate paragraphs, and let _me_ format it
| how I like to read it.
| turtleyacht wrote:
| By the way, HN will flow jagged text when it's not prepended
| with two spaces (code markup). Had to mark it up on purpose.
|
| So, you wouldn't know this sentence was separated into six
| lines. But in editing, it preserves the literal format.
| sepositus wrote:
| I just had an interesting experience this morning with my oldest
| child. He wanted to write a letter to the Splatoon development
| team to show his appreciation for the game. He printed it out and
| brought it to me with a proud expression on his face. It took
| about five seconds to catch the AI smell.
|
| We then proceeded to have a conversation about how some people
| might feel if an appreciation letter was given to them that was
| clearly written by AI. I had to explain how it feels sort of cold
| and impersonal which leads to losing the effect he's looking to
| have.
|
| To be honest, though, it really got me thinking about what the
| future is going to look like. What do I know? Maybe people will
| just stop caring about the human touch. It seems like a massive
| loss to me, but I'm also getting older.
|
| I let him send the letter anyways. Maybe in an ironic twist an AI
| will respond to it.
| xandrius wrote:
| Handwritten letters are still better than typed. That's why we
| still get authors to actually sign their book and not just put
| a print of it.
|
| So, no, there is no evidence that AI will change stuff. We had
| canned responses and template answers for a long time but
| people still like talking to a real human being.
|
| P.S. I think you should have told them to write a thank you
| letter themselves as a fun game to compare with the AI one and
| send that one instead.
| klabb3 wrote:
| > Maybe people will just stop caring about the human touch.
|
| No they won't. This has happened many times in the past already
| with live theatre -> movies, live music -> radio etc. They
| don't replace, but break into different categories where the
| new thing is cheap and abundant. When a corporation writes you
| a shit letter with "We miss you, Derek" all reasonable people
| know what they're looking at.
|
| Look, it's about basic economics. It doesn't matter how "good"
| the generated song for someone's birthday is. What matters is
| the time, money and effort. In some cases writing a prompt for
| one-time use is not bad. If you're generating individual output
| at scale without any human attention, nobody will appreciate
| the "gesture".
|
| What bothers me is the content farm for X shit-startups and
| tech-cos thinking they're replacing humans _without_ side
| effects. It'll work just as well as those fake signatures by
| the CEO in junk mail: it'll deceive only for a short time, and
| maybe older people who may be permanently screwed. It'll just
| be yet another comms channel saturated with spam, which is
| entirely fungible from the heaps of all other spam. A classic
| race to the bottom.
| jraph wrote:
| > Maybe people will just stop caring about the human touch
|
| I doubt it. We human beings seem intrinsically motivated and
| enthusiastic about human connections. I believe we are _wired_
| like this. I know things changesbut I would need some strong
| evidence before even playing with the idea that we 'll stop
| caring about the human touch.
|
| Now, as much as I hate AI, that doesn't necessarily mean AI-
| free. Or even handwritten. It just needs to be _some_ human
| touch. I would enjoy a handwritten letter but wouldn 't mind an
| email at all. But maybe someone else would find it lazy and
| tasteless, just as I would find an ai generated text lazy and
| tasteless.
|
| Maybe the prompt you can guess the person who sent you some ai-
| generated text used can already be perceived as _some_ human
| touch. Maybe there is a threshold though.
|
| Now, could it be that your child wanted to impress _you_ with a
| perfectly written letter, or even with their ai prompt
| mastering?
|
| Anyway, good anecdote, good perspective, good for you to have
| had the conversation and let them proceed anyway. Thanks for
| sharing.
| sepositus wrote:
| > I believe we are wired like this
|
| Absolutely, and I think because of this we'll never see the
| desire go away completely. However, I'm imagining some
| dystopian future where human touch is so rare that people
| _forget_ how much it means to them. It's like scrolling
| through the endless slop of Netflix and then coming to some
| rare gem of a film where you're reminded what genuine art is.
| mulmen wrote:
| I'm glad you let him send his message.
|
| AI is the future whether you like it or not. Teaching him to
| use that tool effectively will serve him far better than
| shaming him for engaging the world in a way you find
| uncomfortable but is acceptable to society.
|
| Consider if you would prefer he write the letter by hand to
| give the script that literal human touch. If not why is it ok
| for the computer to make the letters but not the words?
|
| In this case the meaningful gesture is sending the message at
| all. He asked the AI to do a thing. That was his idea. AI just
| did the boring work of making it palatable to humans.
|
| Much like driving and everything else automation takes away
| writing is something most people are profoundly bad at. Nothing
| is lost when an AI generates a message a human requested.
| A4ET8a8uTh0_v2 wrote:
| It is interesting. It is interesting in several different ways
| too:
|
| - The timing is interesting as Altman opened US branches of his
| 'prove humanness' project that hides the biometrics gathering
| effort - The problem is interesting, because on HN alone, the
| weight of traffic from various AI bots seems to have become a
| known ( and reported ) issue - The fact that some still need to
| be convinced of this need ( it is a real need, but as first point
| notes, there are clear winners of some of the proposals out there
| ) resulting in articles like these - Interesting second and third
| order impact on society mentioned ( and not mentioned ) in the
| article
|
| Like I said. Interesting.
| klabb3 wrote:
| It's futile. The first thing people will do if the "written by
| a human" crypto-signature takes off is to wire it up with LLMs.
| You can't make any reasonable guarantees on authorship unless
| you change the entire hardware stack and tack on loads of DRM.
|
| Even if that happens, and say Apple integrates sigs all the way
| down through their system UI keyboards, secure enclaves, and
| TPM, you think they're going to conform to some shitcoin spec?
| Nah man, they'll use their own.
| pixl97 wrote:
| >You can't make any reasonable guarantees on authorship
| unless you change the entire hardware stack and tack on loads
| of DRM.
|
| Even then you can't trust it. Companies write DRM and tend to
| have actual humans run the place. If the government where
| these humans live decides to point guns at them and demand
| access, most humans are going to give up the key before they
| give up their life.
| A4ET8a8uTh0_v2 wrote:
| Oddly, this is the weirdest thing about all the 'its llms all
| the say down'. The problem starts with user trusting the
| provider to be dumb pipes. The moment they become smart pipes
| ( and one could argue we are there already ), all bets are
| off, because you have no comfort that even if you sent the
| message, it was not altered by an overzealous company bot.
|
| Edit:
|
| Us tomorrow: Your honor, my device clearly shows timestamps
| and allegedly offending message, but you will note the
| discrepancy in style, time and message that suggests that is
| not the case.
|
| LLM judge: Guilty.
|
| Edit2:
|
| Amusingly, the problem has been a human problem all along.
| throwaway173738 wrote:
| I love how Altman created the problem and is also selling a
| solution. "Getting you coming and going" as it were.
| michaeljx wrote:
| I "prompt prong", i.e pass all the AI emails that I receive
| through an AI and ask it to write a response. At what point do we
| get the 2 AIs to email each other directly, without as bio-agents
| having to pretend that we wrote/read them?
| rorylaitila wrote:
| It's becoming a big problem. But also, pretty easy now to filter
| out authentic people. In terms of AI in personal and business
| relationships, the maximum I'll do is short auto complete.
| Superhuman email, which I use, just came out with AI written
| replies, and it is too much.
|
| I simply don't want to live in a world where all I am doing is
| talking back and forth with AI that are pretending to be people.
| What is the point in that? I am working on https://humancrm.io
| and I am explicitly not putting any AI into it. Hopefully there
| are more than a few people that feel like I do.
| xandrius wrote:
| Big problem? Who are you talking to?
|
| Really, it someone can't even type a response back, were you
| ever close to begin with? Unless they always had some level of
| anxiety when sending you a text, in that case it's good for
| them to still interact with you without feeling the negative
| effects (some people truly have issues sending even the
| simplest messages back to friends).
|
| But in reality no, this won't be a problem. We had copy paste
| and template system for decades and nobody are using those. And
| at the end of the day, even if our AI plan our meeting for us,
| and then end up meeting IRL, what's the problem?
| throwaway173738 wrote:
| If you have that kind of anxiety you either need different
| friends or you need therapy.
| OJFord wrote:
| > Big problem? Who are you talking to?
|
| Maybe they work for this guy (or someone like them):
| https://news.ycombinator.com/item?id=43861328
| rorylaitila wrote:
| I should have added the context "it's a big problem in
| relation to sales" which is becoming quite an arms race to
| ever greater automation and "personalization" one-upsmanship.
| There is a lot of FOMO and not automating. In the long run I
| don't think most of the sales automation will actually work
| the way people think, so in that sense, it won't be a
| problem.
| jsemrau wrote:
| I thought same about paying in cash vs. touchpay. It's just
| more human, I can take time to chat with the person. Fast
| forward to 2025, I prefer touch and go because its just more
| efficient.
|
| I think most business email conversations will follow the same
| route. We don't need all of this bla bla. Just "here is the
| information you need" -> "update me if you (a) need more info
| or (b) if the task is done"
|
| What is notable here is though that we continue to reduce
| human/human interactions and that will eventually lead to a de-
| sensitizing of human culture.
| rorylaitila wrote:
| Yeah I am fine with efficiency. I guess its a matter of where
| you draw the line. In terms of my person interactions, I draw
| the line at autocomplete. If the valuable input from the
| person on the other end is simply to confirm and hit send on
| whatever the AI produces, might as well as remove their role
| entirely and let me solve the problem myself.
| mulmen wrote:
| I use generative AI in the exact opposite way. I write the
| simple facts and let GenAI make it palatable to humans.
| doctorhandshake wrote:
| Ironically (and perhaps proving the author's point), when reading
| I couldn't help feeling like this was at least AI-assisted
| writing, if not prasted directly.
| jstanley wrote:
| A big giveaway is that for most of the post the author uses
| hyphens instead of em-dashes, obviously because em-dashes are
| too difficult to type, but then uses em-dashes in a handful of
| places that sound exactly like the kind of thing ChatGPT would
| say.
| dotslashmain wrote:
| Some people (like the author) are going to seriously overthink
| this at first, and then we'll get used to it.
| rpgwaiter wrote:
| Idk why you'd want to get used to it. I'm very lucky to have a
| job that doesn't mandate AI use, and as far as I can tell I
| haven't been hit by any work AI emails. My social media bubble
| on Mastodon is extremely anti-AI, I pretty much never have to
| deal with slop.
|
| On the rare occasion I see some GPT garbage, I either block the
| sender, or if I know a human is involved I explain how
| insulting it is and let them know they're one slop message away
| from blocked.
|
| Getting used to it is a surefire way to make your communication
| experience much worse.
| Terr_ wrote:
| > That's when it dawned on me: we don't have a vocabulary for
| this.
|
| I'd like to highlight the words "counterfeiting" and
| "debasement", as vocabulary that could apply to the underlying
| cause of these interactions. To recycle an old comment [0]:
|
| > Yeah, one of their most "effective" uses [of LLMs] is to
| counterfeit signals that we have relied on--wisely or not--to
| estimate deeper practical truths. Stuff like "did this person
| invest some time into this" or "does this person have knowledge
| of a field" or "can they even think straight."
|
| > Oh, sure, qualitatively speaking it's not new, people could
| have used form-letters, hired a ghostwriter, or simply sank time
| and effort into a good lie... but the quantitative change of
| "Bot, write something that appears heartfelt and clever" is huge.
|
| > In some cases that's devastating--like trying to avert
| botting/sockpuppet operations online--and in others we might have
| to cope by saying stuff like: "Fuck it, personal essays and cover
| letters are meaningless now, just put down the raw bullet-
| points."
|
| [0] https://news.ycombinator.com/item?id=41675602
| SoftTalker wrote:
| I like _chatshit_ --- bullshit, but written by AI.
| mcherm wrote:
| Oh, now THAT one seems likely to actually catch on!
| AnimalMuppet wrote:
| Debasement (in the currency sense). That's _exactly_ what it
| is.
|
| And then you get to Gresham's Law: "Bad money drives out good"
| (that is, drives it out of circulation)...
| ash_091 wrote:
| Last week I got an email from a manager about the number of free
| beverage taps in our office being reduced.
|
| They'd clearly dumped the memo they got about the reduction into
| some AI with a prompt to write a "fun" announcement. The result
| was a mess of weird AI positivity and a fun "fact" which was so
| ludicrous that the manager can't have read it before sending.
|
| I don't mind reading stuff that has been written with assistance
| from AI, but I definitely think it's concerning that people are
| apparently willing to send purely AI generated copy without at
| least reviewing for correctness and tone first.
| pizzafeelsright wrote:
| Tone policing? I'm fine with it although I too got an email
| from corporate about an event with the same type of fun energy.
| My impression changed of the event as it was no longer personal
| but mechanical.
|
| There's always been some innate ability to recognize effort and
| experience. I don't know the word for it but looking at a
| child's or experienced artist drawing you just know if they put
| in minimal or extra effort.
| ahowardm wrote:
| In the era of AI every time I have to read a report I ask myself:
| if the author probably didn't spend his time writing this, why
| should I waste mine reading it?
| et1337 wrote:
| I DM'd someone a specific question recently and got back some
| generic, long-winded, impeccably formatted bullet points when
| previously this person had barely used any punctuation. I said
| something like, wow I like your formatting! Is that an extension
| or something? I always have a hard time with Slack formatting,
| etc. They replied that they were "trying something called
| structured communication."
|
| It probably made me angrier than it should have. Now I'm
| wondering if I'm the "old man yelling at cloud".
| noman-land wrote:
| I hope you told this person that their experiment upset you.
| et1337 wrote:
| Maybe I should have been more direct, but this was someone at
| work that I don't speak to that often.
| noman-land wrote:
| Yeah but they told you they were doing an experiment. Help
| them with their experiment.
| analog31 wrote:
| I consider an AI response to a question, to be the new form of
| "let me Google that for you."
| miragecraft wrote:
| This is just people being lazy and asking AI to do their
| proverbial homework instead of using it as a teacher to help them
| improve.
| agentultra wrote:
| Makes me wonder why people think I want to read what they send me
| if they haven't even bothered to write it.
| Der_Einzige wrote:
| Sorry, but any trick you think you have for detecting AI
| generated text is defeated by high temperature sampling (which
| works now with good samplers like min_p/top-n sigma) and the
| anti-slop sampler: https://github.com/sam-paech/antislop-sampler
|
| You'll be able to detect someone running base ChatGPT or
| something - but even base ChatGPT with a temperature of 2 has a
| very, very different response style - and that's before you
| simply get creative with the system prompt.
|
| And yes, it's trivial to get the model to not use em dashes, or
| the wrong kind of quotes, or any other tell you think you have
| against it.
| analog31 wrote:
| Even before the advent of AI, I already realized that most of
| what's written by people isn't worth reading. This is why I don't
| read most of the e-mails that I receive.
|
| Before AI, it was hard for many people to write literate text. I
| was OK with that, if the text was worth reading. I don't need to
| be entertained, just informed.
|
| The thing that gets me about AI is not that what it generates is
| un-original, but if it's trained on the bulk of human text, then
| what it generates is _not worth reading_.
| jgalt212 wrote:
| I guess this is effective way to let the recipient know how
| important they are to you. AI Ghosting, if you will.
___________________________________________________________________
(page generated 2025-05-03 23:02 UTC)