Join the discussion
Write your take first — we'll ask for email only when you're ready to publish.
- Hacker News
- I don’t know. As a neurodivergent person I have been insulted for my entire life for lacking “communication skills” so I’m glad there is something for levelling the playing field.by charlieyu1
- Your bad, human, prose is a hundred times better than any chatgpt slop. Mistakes and all (also grammar and spelling was already largely a solved problem).by arresin
- Hear hear. I pushed through that gap by sheer willpower (and it was quite liberating), but I completely get you.by rcarmo
- I’d rather be insulted for something I am and can at least try to improve, than praised for something I’m not or can’t do, despite my physiological shortcomings.
- It only levels the field between you and a million spambots, which arguably makes you look even worse than before.
- People at work have fed me obviously AI generated documentation and blogposts. I've gotten to the point where I can make fairly accurate guesses as to which model generated it. I've started to just reject them because the alternative is getting told to rewrite them to "not look AI".by xena
- I don't like reading content that has not been generated with care. The use of LLMs is largely orthogonal to that. If a non-native English speaker uses an LLM to craft a response so I can consume it, that's great. As long as there is care, I don't mind the source.by doug_durham
- I think it is too late. There is non zero profit of people visiting your content, and there is close to zero cost to make it. It is the same problem with music, in fact I search youtube music only with before:2022.
I recently wrote about the dead internet https://punkx.org/jackdoe/zero.txt out of frustration.
I used to fight against it, I thought we should do "proof of humanity", or create rings of trust for humans, but now I think the ship has sailed.
Today a colleague was sharing their screen on google docs and a big "USE GEMINI AI TO WRITE THE DOCUMENT" button was front and center. I am fairly certain that by end of year most words you read will be tokens.
I am working towards moving my pi-hole from blacklist to whitelist, and after that just using local indexes with some datahorading. (squid, wikipedia, SO, rfcs, libc, kernel.git etc)
Maybe in the future we just exchange local copies of our local "internet" via sdcards, like in Cuba's Sneakernet[1] El Paquete Semenal[2].
by jackdoe - Uhh, that's a lot of links: https://download.kiwix.org/zim/wikipedia/
Where are the explanations what all of them mean? What is (nothing) vs `maxi` vs `mini` vs `nopic`? What is `100` vs `all` vs `top1m` vs `top` vs `wp1-0.8`?
by tasuki - > thought we should do "proof of humanity"
I thought about this in another context and then I realized: what system is going to declare you're human or not? AI of course
by gosub100 - Recently I had to give one of my vendors a dressing down about LLM use in emails. He was sending me these ridiculous emails where the LLM was going off the rails suggesting all sorts of features etc that were exploding the scope of the project. I told him he needs to just send the bullet notes next time instead of pasting those into ChatGPT and pasting the output into an email.
- I was shouting to my friend and partner the other day, that he is absolutely to ever stop sending me LLM-generated mails, even if the best he can come with is full of punctuation and grammar errors.by larodi
- I don't get all this complaining, TBH. I have been blogging for over 25 years (20+ on the same site), been using em dashes ever since I switched to a Mac (and because the Markdown parser I use converts double dashes to it, which I quite like when I'm banging out text in vim), and have made it a point of running long-form posts through an LLM asking it to critique my text for readability because I have a tendency for very long sentences/passages.
AI is a tool to help you _finish_ stuff, like a wood sander. It's not something you should use as a hacksaw, or as a hammer. As long as you are writing with your own voice, it's just better autocorrect.
by rcarmo - 100% agree. Using it to polish your sentences or fix small grammar/syntax issues is a great use case in my opinion. I specifically ask it not to completely rewrite or change my voice.
It can also double as a peer reviewer and point out potential counterarguments, so you can address them upfront.
- The complaint is because people (marketers and people marketing themselves) are not using it that way, and instead are using it to generate low value blogspam.by fullshark
- The problem is that a lot of people use it for a whole lot more than just polish. The LLM voice in a text get quite jarring very quickly.by yxhuvud
- If they can’t be bothered to write it, why should I be bothered to read it?by icapybara
- Tired meme. If you can't be bothered to think up an original idea, why bother to post?by CuriouslyC
- Because the author has something to say and needs help saying it?
pre-AI scientists would publish papers and then journalists would write summaries which were usually misleading and often wrong.
An AI operating on its own would likely be no better than the journalist, but an AI supervised by the original scientist quite likely might do a better job.
by bryanlarsen - Now that I think about it, it's rather ironic that's a quote because you didn't write it.by AlienRobot
- They used to say judge the message, not the messenger.
But you are saying that is wrong, you should judge the messenger, not the message.
by dist-epoch - Many of those who can't be bothered to write what they publish probably can't be bothered to read it themselves, either. Not by humans and certainly not for humans.by alxmdev
- > If they can’t be bothered to write it, why should I be bothered to read it?
Isn't that the same with AI-generated source code? If lazy programmers didn't bother writing it, why should I bother reading it? I'll ask the AI to understand it and to make the necessary changes. Now, let's repeat this process over and over. I wonder what would be the state of such code over time. We are clearly walking this path.
by thw_9a83c - I'm sure lots of "readers" of such articles fed it to another AI model to summarize it, thereby completely bypassing the usual human experience of writing and then careful (and critical) reading and parsing of the article text. I weep for the future.
Also, reminds me of this cartoon from March 2023. [0]
[0] https://marketoonist.com/2023/03/ai-written-ai-read.html
by abixb - > No, don't use it to fix your grammar, or for translations
I think that's the best use case and it's not AI related as spell-checkers and translation integrations exist forever, now they are just better.
Especially for non-native speakers that work in a globalized market. Why wouldn't they use the tool in their toolbox?
by dewey - > Especially for non-native speakers that work in a globalized market. Why wouldn't they use the tool in their toolbox?
My wife is ESL. She's asked me to review documents such as her resume, emails, etc. It's immediately obvious to me that it's been run through ChatGPT, and I'm sure it's immediately obvious to whomever she's sending the email. While it's a great tool to suggest alternatives and fix grammar mistakes that Word etc don't catch, using it wholesale to generate text is so obvious, you may as well write "yo unc gimme a job rn fr no cap" and your odds of impressing a recruiter would be about the same. (the latter might actually be better since it helps you stand out.)
Humans are really good at pattern matching, even unconsciously. When ChatGPT first came out people here were freaking out about how human it sounded. Yet by now most people have a strong intuition for what sounds ChatGPT-generated, and if you paste a GPT-generated comment here you'll (rightfully) get downvoted and flagged to oblivion.
So why wouldn't you use it? Because it masks the authenticity in your writing, at a time when authenticity is at a premium.
by mjr00 - Yah, it is very strange to equivocate using AI as a spell checker and a whole AI written article. Being charitable, they meant asking the AI re-write your whole post, rather than just using it to suggest comma placement, but as written the article seems to suggest a blog post with grammar errors is more Human™ than one without.by boscillator
- Because it doesn’t just fix your grammar, it makes you sound suspiciously like spam.by j4yav
- I just hit the back button as soon as my "this feels like AI" sense tingles.
Now you could argue but you don't know it was AI it could just be really mediocre writing - it could indeed but I hit the back button there as well so it's a wash either way.
by noir_lord - I do the same, but for blog posts complaining about AI.
At this point, I don't know there's much more to be said on the topic. Lines of contention are drawn, and all that's left is to see what people decide to do.
by shadowgovt