11–20 of 262 posts
How is the author complaining about the quality of their own writing while admitting to not even bothering reading what they wrote, let alone editing it?
(Also, why would using a LLM based grammar checker trigger an AI writing detector? Did it end up rewriting substantial parts of the original submission?)
> This post, is written without any tools assistance I just wrote what my brain is instructing to type (might not reread it before posting). How is the author complaining about the quality of their own writing while admitting to not even bothering reading what they wrote, let alone editing it? (Also, why would using a LLM based grammar checker trigger an AI writing detector? Did it end up rewriting substantial parts…
> This post, is written without any tools assistance I just wrote what my brain is instructing to type (might not reread it before posting). How is the author complaining about the quality of their own writing while admitting to not even bothering reading what they wrote, let alone editing it? (Also, why would using a LLM based grammar checker trigger an AI writing detector? Did it end up rewriting substantial parts…
> This post, is written without any tools assistance I just wrote what my brain is instructing to type (might not reread it before posting). How is the author complaining about the quality of their own writing while admitting to not even bothering reading what they wrote, let alone editing it? (Also, why would using a LLM based grammar checker trigger an AI writing detector? Did it end up rewriting substantial parts…
I get that the mainstream ones have been RLHF'd to death, but surely there must be others that are capable?
1. There was a lot of slop pre-AI. In fact I’d say the majority of published writing was bad, formulaic, and just written to manipulate your emotions. So in some sense, I don’t really think pre-AI slop had more value. It’s just cheaper to make now.
2. AI has prompted me to study more off-beat writers that followed the rules of language a little less frequently. This includes a lot of people from circa 1890-1970, when experimenting with form was really in vogue.
3. Which brings me to my third point, which is that no matter how much the AI actually knows about writing, the person prompting it is limited by their own education and knowledge of writers. You can’t say, “make me a post in the style of Burroughs” if you don’t know who Burroughs was, or what his writing style was. So in a sense there is an increased importance to being educated about writing itself. Without it you’re limited in your ability to use AIs to write stuff and in your awareness of how much your non-AI written work is influenced by AI writing.
I never passed any AI writing as my own. I would feel utterly awful. Also, I love tweaking words until they sound perfect.
The number of people who just nonchalantly admit that AI writes their messages is honestly scaring me.
This is exactly same struggle for me. Writing technical content about PostgreSQL and balancing my voice without sounding like LLM written is genuinely difficult. As English is not my first language, I do run into problem where the line between fix my clumsy sentence and rewrite my thought is very thin. Same with writing "boring" technical explanation and more approachable content. I'm getting pushed back for both.
Earlier quoted context omitted.
Don't want to sound like an llm? Don't read llm content. Remove yourself from places where you might be liable to read it.
It's not that simple. LLMs were trained on lots of writing, and the "LLM voice" resembles in many ways good English prose, or at least effective public communications voice. For years, even before LLMs, there have been trends of varied popularity to, for lack of a better word, regress - intentionally omitting capitalization, punctuation, or other important details which convey meaning. I rejected those, and likewise…
That should leave you with media sources like nyt and your local library, which seems healthier to me. And maybe it might encourage a new type of forum to emerge where there is some decentralized vetting that you are a human, like verifying by inputting the random hash posted outside the local maker space.
> "..but maybe it's a good thing that most of us don't allow this technology to reframe our thoughts."
No, you're not the only one experiencing this: I too had the same concerns as you: with every new thought, every new creation, I had to ask the AI's opinion, as if I were no longer able to judge, to decide, without consulting the AI (...just to be safe, you never know...).
The only way to regain your creative ability is to write down your thoughts yourself, read, reread, rewrite, correct, express your opinion...
What AI can't do is convey emotions.