Among you, the HN-community who writes, how much do you use AI in the process?
There is somewhat a spectrum from none, to grammarly-style cleanup, to whisprflow to 'look at my project, write me a HN-ready post'.
Personally, I've done a 180, where I started at light touchups to rough draft-to-polished back to almost none, to keep the writing my own (I do some writing at jakobs.dev).
As a reader, do you believe you notice, and care?
Every word I write is my own, for better or worse. Every line of code, too. I started blogging again just to have a means of expression. No one reads it. No one cares. It isn't optmimized for SEO or whatever, but it's mine.
It's weird and sad that it's come to the point, after just a few years, that that's a thing. I'm just doing what I've always been doing but now people will call you a Luddite and a mental defective for not optimizing every aspect of your life around a chatbot, even if it sucks the fun out of everything.
1. Introduction
2. Transition sentences
3. Middle paragraphs
4. Clean conclusions
which I find doesn't vibe with my style of writing where I tend to meander quite a bit then try to wrap things up back in a different manner
To bounce a draft of something off an LLM and get a critique? Frequently. But I include in my prompt to never re-write or suggest improvements, and to solely offer critique. I do my own fixes.
I have used AI for medical research and that was a mistake. It will leave out potential risks if the number of people at risk are lower than some percentage. To get real risks one has to already know the risks and tease it out of the AI then suddenly it "knows".
But I don't like having AI do any of my communication oriented writing. Unless it's technical documentation about something the AI wrote, but even then I usually am properly quoting the AI in my own writing. Not parading it's ideas as my own.
I feel like it defeats the purpose of me trying to communicate my ideas to people. My ideas then get tainted by the AI's knowledge when I use it to produce text for me. Also, I'm a very bad writer and want to improve on that front, so me writing more can help me improve.
Overly optimized writing now triggers the same feeling as overly optimized SEO pages used to.
(in the case of writing,) AI often cant meaningfully increase the information density of output text relative to that of the input text , but its great for summarization and some synthesis.
if you give it a short prompt to write a long essay, the essay wont be that good.For me, the point of taking and posting the notes is to solidify the lecture in my mind, think about it, etc. Using AI for too much of this process would defeat the purpose.
Am I using british english consistently? Are there any grammatical errors? Etc.
I still feel it's important to fix whatever it finds by hand though.
I personally find it okay / convenient to have AI respond to PR review comments with respect to something being addressed or why it was not. That text is often pretty mechanical.
I like to “talk to” LLMs, yet I never ever use them in my writings. Not even to proof.
Discussing ideas is insightful as LLMs are actually compendiums of whatever may have been made available before, as found in training sets.
As far as what I read, most of what anyone writes is a varying degree of slop to the proficient reader. First I skim, and gauge information density. Most published content aims to develop word count, exercise the author’s idiosyncrasies, and then provide useful or insightful detail (while I’m sure it begins in the opposite order, the public product usually ends up in reverse.)
Typically human writers bury their point after long winded meandering, often pretending the reader has never heard of or considered the most basic developing ideas.
LLMs like to iterate, itemize, and propose every varying nuance unnecessarily.
I enjoy writing which thoughtfully preempts the audience. Delivering the whole point early on, and then drifts into worth while conjectures or details. This indicates the author values my time and honestly has something worth while to share.
Unless it is purely for enjoyment, such as fiction, in which case build ups and nuanced twists are pleasing.
Beware the itemized and iterative diatribe, for those are the works of mechanical compilation!
My thoughts on this are: if you are soulless I don't care what you write about and how you communicate, but if you try to present human, personal ideas with heavy AI writing signs I will give you the same consideration as if a toaster was talking to me.
For example: any kind of corporation communication, linkeding, marketing, pure technical docs, code, etc. I don't care the slightest, it never was human communication, they are just artifacts. I don't care if it's slop, I'm ok talking to your claw slack bot if when I ask I get the massaged info I need.
But if you trick me into talking with you/reading your blog and you outsource your thinking and/or writing to a clanker without disclosing it or convincing me why, you are silicon to me.
It does not help that their work gave them a major promotion for being so pro-AI adoption.
There’s a sickness that AI brings, and the cost to everyone is under appreciated. There’s benefits too, but the validation loop is like a poison, and it seems especially potent for management types.
i want to be respectful of their time and when i write genuinely sometimes i write too much.
i like using AI to brainstorm how can i take what i wrote (maybe 500 characters) and make it concise (like 300 characters)
then i try to manually stitch it together
