The most sensible solution that I've seen so far is to delineate human spaces from LLM spaces. Technical mumbo-jumbo can be LLM as an artifact, and human thought can just point to them.

A good code or commit comment should succinctly explain the subtle details or motivation not obvious from the content itself. AI comments tend to wax poetic about surface details.

The same goes for articles. I want the author to distill the material using their own voice and hard earned expertise. I don't want a generic text which sort of tells the tale.

> I would much rather someone be themselves and write with their own voice, or with some passion

I recently had a fairly successful article here on HN. The common thread among comments was appreciation for the passion and it being "hand-crafted". I am thrilled my passion shone through despite the dry subject. Still, it's sad that something simply being written by a human has become such a rare compliment.

My take here is, I think the real issue is AI pretending to be human by mocking a human style.

I'd always rather read an actual person's thoughts, clearly communicated and written down in their voice. But, there's clearly a subsection of people who don't want to (or aren't confident in / etc) writing their ideas fully.

If they have a bullet point list, and they want to use AI to neaten up the bullet point list and present it back, that's fine by me. I wouldn't have minded reading their thoughts in a messy list, but AI is serving to make it more readable.

What I think everyone hates, is AI being used to convert the list into pseudo-thought-out-prose. It just contains a bunch of bloat that wastes people's time.

So my advice if you use AI to write and don't want to change: keep going, just stop seeing the desired output as human prose. Just neaten up your bullet point list / promot / whatever, and send me that.

> Reading a document like this isn’t just difficult—it is punishing.

I hope this was an in-joke but my fingers still reflexively closed the tab as soon as I saw it and I had to go reopen it again…

We're at a point where I assume that whatever I read is something the "author" endorses, but may not have written. And that may be more important than the specific words an author may have chosen themselves. I doubt we can ever go back.

What should worry us is that AI may introduce a slant that the prompter assimilates during their conversation with the tool, which could introduce a subtle control lever over public opinion.

“Language can become a screen which stands between the thinker and reality. This is the reason why true creativity often starts where language ends.” — Arthur Koestler

We do not experience the world as words. Words and language are already an hallucination on top of reality and LLMs are an hallucination on top of language.

There are many use cases for LLMs that will help humanity progress. Putting AI in between yourself and the people you want to build productive relationships with is not one of them.

I think AI can write good already. It’s all up to the prompt. Strong guidelines. Without them, it will sound like an AI because, just as the article said, the language reflects the writer.

It’s funny, this pattern is very similar to how I write code with AI these days.

I write almost every line but I have one or more agents always running against the changes alongside me evaluating correctness, suggesting approaches, auditing test coverage, etc.

I’ll also use it for grunt work refactors, scaffolding, resolving merge/rebase conflicts, etc.

> I also see more and more pull request summaries clearly written by machines for machines. They are rich in detail: this was changed to that, these things were split, those things were merged, this was left untouched, tests were added for this, and so on. But I’m left asking: Why are we doing this? What is the value? How risky or urgent is this work? Where do you want my input? What should I pay attention to?

My own solution for this (for now at least) is to add a preamble section “AI;DR” to PR descriptions, filling what a TL;DR section would have done in the past. I’ve socialized to my peers that no part of such a section will be written by AI, and it should contain exactly this missing high-level content described by the author.