Write as yourself. You’re being paid for your expertise and insights. Communicate them directly to us. Copying and pasting Claude responses into Slack or an email directly shifts the burden of comprehension and understanding to everyone else, and worse, risks skipping that step for yourself. Even if you’re fundamentally using Claude to gather your thoughts or help you prepare a response, you need to be writing it yourself, in your own voice. Not having Claude ape your voice, or “make it sound less like AI”. You, directly. Doing this will further reinforce your own understanding of the state of things, the same way teaching someone is the best way to learn. As a guideline: for Slack and email comms, this should be near-universally written as you. For deliverables that are longer form and follow a template like proposals, roadmap/discovery work, etc., use of agents is expected but, see Principle #2. (Own the Output.)
As such, isn't the question really about
1) whether it was communicated clearly/effectively 2) whether it should have been communicated to others at all (rather than something that you just dealt with yourself)
As for the writer being paid to use their expertise, learn more things etc: absolutely. But they could spend whatever time is required to think about the issue and then still have Ai write it at their direction. And, regardless of who wrote it, it could still be poor, irrelevant or misleading.
I'm not defending AI over-usage, but you folks evidently haven't thought about any of this nearly sufficiently enough... and the policy doesn't seem to be addressing what it ought to be: whether everyone is actually thinking, learning, discerning, deciding etc...
Ps I passed your comments to chatgpt and asked what it thought about it all, and it missed all of this. But then agreed once I brought it all up. Then it did a poor job of writing it's own response. I couldn't be bothered to guide it towards something more appropriate. https://chatgpt.com/share/6a86cc31-f058-83e8-8abd-3921b518ec...
I think your linked chat demonstrates quite well why you are wrong.
The sender's burden is a few lines of text "Write me $FOO, conclude with $BAR".
The recipients burden is an order of magnitude greater, trying to find your conclusion in hundreds of lines of text.
After all, your initial line:
> But isn't that "burden" on others always going to be there, whether you or AI wrote it? That's what communicating is...
Is concise and clearly understandable in 21 words . I understand your point even if I don't agree with it.
The linked chat has 1424 words. Your point is in there somewhere, but now you want the recipient to play "Where's Waldo" to find your point.
(BTW: I can't believe just how effectively you argued against your own point! I have never seen someone do it as effectively as this in all the time I have been reading HN)
You say that my initial line is good, but then that you disagreed with it without saying why/making a counter argument.
As for sending anyone on a where's Waldo adventure, you've fundamentally missed the point, which was completely self-contained in my comment - the postscript link to chatgpt was just, at best, reference material for anyone who cared to look.
As for the fact that chatgpt did a terrible job of it all, that doesn't undermine my point even slightly. Because my point, made very clearly in my post, was that you can't just offload everything to AI. You have to, at the very least, give it context, direction etc. At that point, it can be quite useful. I use it effectively all the time to polish my writing, remove my worst tendencies etc.
If I cared enough, I'd have had it polish this response to make it significantly less aggressive while making the same point. But you don't deserve that, because you didn't put in even the slightest effort to communicate in good faith
Not the same burden.
The value of the person is digesting and giving the materially important stuff in the answer (including their expertise and opinion). Not the AI diarrhea.
You can also direct the Ai to write about specific points and, again, it may write somehing good or bad. I provided a link to a chat I had with chatgpt about this very conversation and it shows everything I was saying.
The fundamental issue is whether people are taking responsibility to communicate what needs to be communicated, in an effective manner. Whether the final words were authored by Ai or not is largely irrelevant.
Theoritically yes. In practice:
- AI responses I get are walls of written diarrhea (even when they're relevant and take our particular context into account). The one's that don't are from people who write well based on their expertise, and who know to edit AI responses and not just paste them.
- I'd rather get the bad/misleading/not-useful response from the person, rather that the good one they made via AI. This way I know they don't know what they're doing, are bad communicators, depend purely on the AI, and thus their position can be eliminated or replaced by someone cheaper merely relaying to/from the AI.
Evidently the people you are communicating with are not good writers/thinkers. That's the prerequisite for effectively using AI for writing help. Just as its the prerequisite for using it to produce code. If you can't code, you can't get Ai to produce good code.
So the real issue here, perhaps, is "vibe writing" rather than "Ai assisted writing".
> I'd rather get the bad/misleading/not-useful response from the person, rather that the good one they made via AI. This way I know they don't know what they're doing, are bad communicators, depend purely on the AI, and thus their position can be eliminated or replaced by someone cheaper merely relaying to/from the AI.
Agreed. But if people who don't understand the issue are feigning competence via AI then that's a fundamental problem of not just competence across many domains, but also character. As I said in another comment, the policy should be something to the effect of "you're paid for competence, integrity etc. If you don't have that, we don't need you. Take that into account when you use Ai"
I get the feeling this is the whole disconnect. People have high expectations from AI but haven't really been on the receiving end of a massive amount of text/code generated by it.
We recently had someone fired for using AI excessively. Before the firing, whenever anyone on the team sat down to give feedback, the only answer was "but everyone is using AI, we are supposed to use AI, how can it be bad?".
Until their boss sat down and forced them to read the verbal diarrhea and they finally admitted that it was 90% hallucinations and AI-filling-the-blanks.
It's the equivalent of vibe coding vs Ai assisted coding.
Vibe writing vs AI assisted writing.
I don’t expect the other person to copy my response to the same chat and paste the AI response back even without reading it.
Don’t be a meat proxy.
While I agree with your overall distinction, I believe the scenario you portray is so unrealistic that you will never encounter it.
Yes; if I could wrangle Claude into communicating exactly as I would have, the practical difference is of course not observable. But that's never happened, and I've had an enterprise license at work for years.
The value in having a human write it is that whenever something bad/misleading/redundant is introduced, that person will learn from the feedback, but an AI will at best have some memory/skill/prompt nudge for next time that might have a shallow/localized band-aid effect, but will never stop doing similar mistakes elsewhere.
For me, the value of receiving comms from a human is different: it communicates not just the idea itself, but the fact that the human already understands the idea.
The people arguing that round-tripping their thoughts through AI still means that it's their thoughts apparently also believe that the game of telephone always results in perfect comms.
It's not just AI; adding more hops to a route changes the message when each hop is a human too, it's just that we notice it more when the hop is an LLM because of just how different they write compared to humans.
Sorry to say, but this is a skill issue (and not necessarily with using Ai). I've used Ai MANY times to help SIGNIFICANTLY improve things that I've written - from a few sentences to many paragraphs.
As I've said in other comments, it'll rapidly and effectively remove all my worst tendencies. And, likewise, sometimes I'll have it write a first draft when I give it appropriate context/guidance that I can't be bothered to fully curate/synthesize. Then, once it gets that draft, I'll take over and tidy it up. You might say that the "can't be bothered" part is the entire issue, and I'll again say that what matters is the final result - a master chef or chief architect is not doing all the grunt work...
Isn't this contradictory? Why do you need to take over if it's doing such a good job?
There's something profoundly disrespectful about it towards the receiver.
Far better, and frankly perfectly acceptable, to say or have it implied or concluded, that Ai helped with the writing in some way. Still, if you're paying attention, it often shouldnt really be all that detectable because you should be aware of AI's smells and you can edit them away. Though, sometimes I'll leave it as obvious that Ai produced something I shared - eg making a well formatted github issue or whatever. But I'll also make sure to convey that a thinking human was behind it, vetted it etc
Of course the level of moral disengagement that's rife around here generally precludes any recognition of such things, as that might be construed as "harshing the vibe on the innovation, man". So I guess everyone had better just settle in and get used to some serious profound disrespect until this type of thing hits enough of a critical mass that a societal ethics framework begins to manifest around it and not be ignored.
I think it can be OK if AI is used to polish up one’s message or to help structure it. Basically help with grammar, formatting, and wording.
But copy pasting an AIs response to a question? Not at all. People earn a reputation over time. TRUST is earned. An AI cannot earn trust. If you copy paste an AI response verbatim, i need to read everything in your message with a big grain of salt. Imagine doing this all the time as a reader? Exhausting and a waste of others time. If i want to read raw AI output i’ll ask the AI directly, thanks.
If you write something in your own words, or at least synthesize an AI response before sharing it with me, then i know i’m reading something that has at least passed the smell test.
It’s the difference between reviewing a pull request from an intern and a principal software engineer.
Though, I think such a writing partner can only really be useful if you know how to write yourself, which also requires knowing how to think (because writing is really just thinking).
So, again, in the end all that really matters is whether you are effectively communicating what needs to be communicated, Ai-assisted or not. It's quite apparent to other thinking people when someone is not doing that, and that hurts their reputation, undermines trust and ultimately leads to losing your job, connections etc...
So, again, the policy really should just be something to the effect of "you're being paid to help us be effective and correct. If you can't do that, we don't need you. Take that into consideration with regards to the tools that you use.". That's obviously not an adequate policy - Ai could surely be helpful in polishing it! But it most definitely couldn't get to the root of the matter on its own.
There is a world of difference between this and copy-pasting LLM output into slack and email.
You can write the first draft with the LLM, or have the LLM help you do an analysis, look at how code works, and so on. But you need to check for the LLM to be correct and you need to be able to articulate what the LLM found in your own words. If you don't you have just offloaded all the work to someone else: someone else will have to check, verify, investigate what could very well be an hallucination (bad context, random bit flips, whatever the cause doesn't matter). Your effective contribution to the discussion will be negative.
And because most ai's are not concise we blame ai's for faults we don't like in humans either: being longwinding
In fact, my tendency is to be far too long-winded, with lots of perhaps-unnecessary parantheticals etc. Ai is often shockingly useful at reining me in and focusing on the issue at hand, as well as removing my worst tendencies towards meanness etc...
Im a good writer, when I want to be. But having Ai review, edit, write drafts etc undoubtedly makes me a better writer/communicator. But only because I'm actively thinking about it all, rather than just saying "respond to this" and then providing copy pasta.
What you wrote just seems common sense to me, and I'm hopeful it will emerge as such given enough time.
https://www.anthropic.com/learn/claude-for-you
This framework provides a more abstract way of thinking when working with AI systems. Especially valuable to me is the pattern of the 4 D's: Delegation, Description, Discernment, and Diligence, which I have now internalized whenever I use AI systems. It helps me understand my own and others' (mis)use of AI systems a bit better.
Like is your insight that copying and pasting something is not understanding it? Do people seriously not known this?
There is no need to Socrates when writing policy docs. Maybe this is why people ignore reading your actual voice messages that you feel a need to also justify this basic request. Sometimes the human voice thinks too highly of itself also.
And maybe GP doesn't want that.
For purposes of training the AI (enablement firm)?
> Write as yourself.
Unless it's your first day on HN (it's not 2008 account), most people use paragraphs and sentences longer than 2-5 words. Something like "You, directly." in the middle of all that has a certain vibe to it.
> Copying and pasting Claude responses... directly shifts the burden of comprehension
Most definitely.
The problem is we live in a world where people try to pass off copy pastes as their own all the time, many even smart enough to remove the em-dashes now. So I don't think that's going away soon. Your first sentence seems human, the rest doesn't match the tone at all.
Isn't that the very thing you just claimed people don't do?
For simple, impactful sentences, I often use short ones. They work better.
Since people started using Claude now I get full context of everything… might be too much sure, but to be honest over-communicating seems better than under-communicating - sure it’s boring and tedious but that shifts the blockage to me.
Otherwise each of these coaxing sessions is something I have to keep in my head until resolved, which is a load in and off itself.
And even better - I can point Claude to that message and gives me a summary. You might say this is silly because we are paying the LLM tax, but knowing what to share, and then verifying against my situation are two different things.
Sometimes they have checked the wrong thing, sometimes they need guidance, sometimes I need to investigate before I can answer. A full LLM message with context is like “summary of their working context that I can resume on my end” so I don’t waste cycles asking or rechecking etc.
And it is still possible to be a learning experience for both of us since the initial message is the start, we can then talk to each other like humans, both much more in sync than before…
It is kinda ironic but I think it does push things in a better direction. Would I have loved it if the message was human, direct and to the point - obviously, but we are all busy, we got shit to do, and this is a useful “resume” mechanic
I'd argue that it's worse because now you have no idea what the actual original context was, just the AI's hallucination of it.
And I think this is where one key argument on the page itself I think fails:
> The person on the other side has the same tools you do.
Yes, the person on the other side has Claude; but in many cases, their Claude doesn't have the same context as my Claude. Saying, "This thing happened in an area of the code I don't have any expertise in. Below is Claude's analysis, which I'm not qualified to assess" is I think a valid time to copy and paste LLM output. [1]
The other thing to say is that it can be a "LMGTFY" style response; i.e., "Why did you ask me that, when you could have gotten a perfectly good answer from ChatGPT?"
[1] In the particular case I'm thinking about, it was a bug in an ocaml component; I don't know ocaml or need to know it; handing Claude's analysis 'caveat emptor' to the maintianers who do know ocaml I think is more helpful than not.
I’m reading a bunch of game modding discords, and this exactly what happens, and what’s hated. Because Claude is also not qualified to assess it. Not by default. The modder’s Claude might be because of the extra context and steering, but the user’s produces usless slop, instead the actual log, and context what the user did is what’s helpful.
I’d say this (when you don’t understand it yourself) is pretty much the worst time to send AI output.
Quick edit: Yeah, I get it in situations like GP where the user is unhelpful, doesn’t want to learn, doesn’t care, and you also can’t tell them to fuck off. But just because the user gives Claude the context they should have given you in the first place, doesn’t make it good.
> a lot of my colleagues messages are “x broken” or “I do y”
These people had below than average communication skills (or at least they were putting below than average effort communicating with you), so LLMs improved their output.
However, the society kinda expects people can and Should improve themselves over time. The issue is that LLMs improved their output, but not necessarily them.
1. I've experienced both (ai paste slop & ai providing helpful context that would otherwise have been absent) but the former has been a much bigger problem than the purported benefits of the latter
2. before the advent of LLMs, I had fairly strong contrarian opinions on the latter (clueless users asking questions with no detail) that I still hold to this day. A good proxy for my opinions on this is that I think https://nohello.net/ is a crutch for undersocialised (or overworked/near-burnout OR undiagnosed autistic) engineers & should not be normalised. The origin of this sentiment is users entering a quiet IRC room & saying "hello", then leaving & not being around when someone eventually replies: in that context nohello is pragmatic, but taking it beyond that to modern high-frequency realtime chat is an antisocial overstep. Leading with the entire context of your query may have some practical benefits but they're honestly overstated in exchange for killing natural social tendencies towards human dialogue. Similarly, a non-technical user giving you zero info about a problem is an opportunity for conversation & mutual education. It's relationship-building & helps build a foundation for further interactions with this person. It was never a problem in the first place; "solving" it is just another milestone on the road toward de-socialising humanity.
I would love to spend more time and talk to the people with their issues so we can come to a common understanding… but neither have the capacity for this, and now things are getting worse I think.
Maybe someday agents are good enough that we just discuss direction and implementation details, and every “developer” becomes more of an executive in a company, free to chat about more than just work, focusing on building human relationships and leaving “the tech stuff” to their agent teams…
And old geezers like us who actually know how to write ourself out of a for loop would be the premium - wizards holding forgotten knowledge, incomprehensible to mere uninitiated mortals. I wonder what distinctive dress and hair styles will we develop :-D
Context is everything.
Btw, Claude code sent my first accidental AI email yesterday. I guess permissions changed? Before it could only draft. Uhoh, waiting for the onslop.
> Since people started using Claude now I get full context of everything… might be too much sure, but to be honest over-communicating seems better than under-communicating - sure it’s boring and tedious but that shifts the blockage to me.
Yes, but the kind of engineer that used to say "X is broken" will send you some slop with missing context and Claude will try to apply general solutions from its training data rather than from the specific context. As somebody on the Ops side that usually helps devs fixing things here and there, in those cases I'd rather them say "X is broken" and then I ask the question to Claude with all the context I have.
Probably the best solution currently is a shared skill with at least some curated context.
I think that's the key. Generally people would rather not receive slop because the receiver has no idea how much thought the sender invested in the communication.
In your example, the more context the better. I think there's an interesting nuance there.
That's an interesting thought. The message can have metadata that says the sender's spent 1/5/10/60/120 minutes or some other metric of how much thinking over the communication happened. Just need a way to monitor people's thoughts and digitize them to make that happen.
Really? How does Claude gather that context?
"read it all, yes even the bullets. especially the bullets"
if you read one thing from a large wall of text, its gonna be the bullet points, right? thats a big part of what they're for?
"It is the conversational equivalent of forwarding the email."
this is a terrible metaphor. it seems to rely on the idea you would be mad if you sent an email to someone and they forwarded to someone else? but thats useful, theyre literally sending it to someone who is presumably better placed to help you, or raising it to someone more important because they think what youre saying is important
Interesting to note the page also ends with "written by a human, on purpose"
Hypocritical to say the least, given the message.
And the result isn't even very angry at all. It's like mildly peeved internal company communication (also in direct violation of "don't send this to your manager")
DS9 Garak "especially the lies" reference maybe?
But the point is. They had search, so why didn't they think to do the search? They had the same tools as I did. And yet regularly they couldn't think of the answer. I still deal with this with AI today. When you ask a question, maybe you're asking for the interpretation of the question by the person on the other side. And it's that interpretation that goes to the LLM. Or otherwise some filter of how that person thinks about the question differently than you do.
This is exactly what I have been feeling in response to this person. They are not contributing. They are creating work for others by pasting every possible problem they can find, most of which are out of scope or context, and letting the requester sort through them which takes hours if you take it seriously.
Some people do and implement everything he asks. They are going way outside the scope of their ticket and fixing every problem in every file they touch. It is chaotic and not good for the codebase because a lot of the problems are repo wide and should not be fixed in isolation.
He doesn't seem to understand why this might bother people or why it takes them so long to respond. He suggested just pulling the comment into Cursor and just asking Cursor to respond...
At that point we are just straight up vibe coding with no human in the loop.
That being said, Cursor's bug bot does a way better job reviewing, and people just ignore it when it asks for something out of scope because it's not a colleague claiming their code is broken.
Dont be a meat proxy! simple
Unique human prompt + AI output which is useful and does a good job expanding on the prompt topic and which when pasted saves people the time of having to go prompt AI on the same topic (fully reviewed by a person before sharing it, to ensure the content has relevance)
Would be interesting if someone said “I spent $2.50 using llm-council with the three top SOTA models and think this reply makes sense: {paste}”
(But better “not sure, though after spending a few bucks on model APIs, perhaps {human-written analysis} is in the right direction”)
Also like you, I did think on the angry version “they have the same AI you do” isn’t likely to be right in the consumer space since their AI is ChatGPT (free version?!) and yours is Claude or something.
There's responsible AI use (which I think you - and I, and many people here) do, and the "I don't care, I've outsourced my brain to OpenAI" type of people.
Over time, it was impressive to see how much the output of the same prompt would differ between us, and it seems to be heavily influenced by the role of the user. I.e. answers coming from the agent of a Data Engineer, a Backend Engineer and a Domain Researcher would not be the same at all. Some of us use harnessing tools, some not, but regardless it we saw differences.
So, for example, when a backend engineer would send a question to a data engineer, there was *some* added value even when it was a lazy copy & paste.
You're gonna need a stronger language. Why do you think people love AI? If I have to think and god forbid read the response I might as well live in the woods or something
DO NOT FUCKING PASTE AI RESPONSES TO ME, EVER, PERIOD.
When someone tells me they "talked to Claude" they sound retarded. Just tell me you looked into it, and I fucking swear to god if you just took what Claude said verbatim and do not understand the problem then you are a cunt and I will stop working with you.
When someone pastes LLM output in chat I completely ignore it. I will just continue the conversation as if they said nothing and silently mark them as cretinous in my mind.
This is not up for debate, you will talk to me in short well formed sentences and get to the fucking point or you will fuck off
Especially today, I will prefer your slightly funny English to something that feels like AI.
And personally I do Support for people around the world and I often ask people to write in their native language.
Then I can translate myself and if the default ai translation gets something wrong I can add necessary context to improve it - which I can't really do if the only thing I get is the bad ai translation.
Der skal kom et tid hvor du skal tale på en sprog på dit kontor. Uden hjælp. Du har bruge for at øve dig, og folk er alligevel pænere end du tror.
(this being written in a foreign language I am learning is part of the point)
And in my experience, us Brits, and USA folk, we tend to be quite understanding of broken English
If you are getting blocks of llm text perhaps it's because you don't deserve their input and could have satisfied your query by not making it someone else's problem.
idk about that
[edit] to extend the connections The charm of a messaging board like this is bringing "join attention" to a topic, not just sharing separate monologs.
And also, always wondering if something was written by a human or statistically generated could be seen as Epistemic Hypervigilance, leading to stress, and loss of epistemic trust. This could reduces opportunities for taking on new information of feedback, and cause rigidity and isolation.
Almost certainly written by AI
It also ends with "written by a human, on purpose"
This is really useful, I wish more people would understand this: you don't have to have an opinion on everything. In fact, you probably shouldn't have an opinion on most things.
My own take on the AI issue: https://blog.gpkb.org/posts/just-send-me-the-prompt/
I like to think it helped, but likely what helped was that nobody seems to use Deep Research anymore, haha.
and i detest ai slop
so my solution is to speak combined with an ai cleanup like macwhisper or wispr flow of all the uhs and ahs and repetitions or rephrasings etc
you can even use ai to pretend a conversation or interview style
then i have stuff to edit
boom
Disagreed. Bad writing from a chatbot beats worse writing from its user every time - regardless of length.
One reason we have so much bot slop is the attractiveness of bad-writing bots to the many people whose writing is even worse.