Hacker News Reader

> The problem with writing, and something I am noticing lately, is that AI robs your ability to create prose in your style. Even if you use AI just a little bit to touch up your writing - you start being influenced - And over time it adds up - you start writing like AI just a little bit every now and then. It seems to be robbing me of my ability to create original prose.

IMHO the fix is to read more books. Obviously I mean human-written books, and specifically with prose that you appreciate.

That was always the fix, long before we had LLMs.

After a hiatus of a few years, when the internet had me, I've gone back to prioritising reading books once more. It's not only great for my writing, it's also just good for the soul. And it's the absolute antithesis of the infinite scroll.

"What you read and how deeply you read matters almost as much as how you love, work, exercise, vote, practice charity, strive for social justice, cultivate kindness and courtesy, worship if you are capable of worship. The mind is an activity and will decay into dark inertia if not sustained by the sustenance of reading.

The great poems, plays, novels, stories teach us how to go on living, even when submerged under forty fathoms of bother and distress. If you live ninety years you will be a battered survivor. Your own mistakes, accidents, failures at otherness beat you down. Rise up at dawn and read something that matters as soon as you can."

-- Harold Bloom

Not too sure - I never read that many books - and it wasnt an issue at all. I think writing more, without AI, is helping me get back to my pre-AI-stink self.

Sure, but if you’re influenced by how AI writes, it stands to reason that you’d be influenced by a great author too, and that may help improve your writing.

I may be misreading, but I think the earlier post was about the LLM being used to edit (clean up) their writing.

It's quite different than just reading a lot of text from other authors and passively absorbing diction and style. Instead, it means getting detailed reinforcement signals in the context of your own active writing effort.

This can start an ironic reinforcement learning pattern where the writer gets tuned to write in the LLM's "corrected" form. Imagine a traditional professional writer, in a long-term partnership with a particular copy editor who gates their output.

> Strategic thought is the other thing

I feel like LLMs can be very helpful to double-check a bunch of stuff here, and evaluate alternative-paths and so on, but the "direction" really has to come from you, then you can use the LLM to double-check and what not.

But if you let the LLM set the "direction", then all bets are off, as they don't have a single care about anything in the world, so they'll just do whatever essentially.

"Direction" is what I mean by strategic thought - but even if the high level direction came from you - the tactical approach might fall prey to the same over confidence which destroys direction setting in the first place.

I use it very sparingly for stuff like this - and even when I do - I discount its outputs heavily and maybe just use it let me see some blind spots and sanity testing.

I've tried teaching AI to copy a writer's style and it fails miserably. And the output out of it always have the same writing style - LLMesque. I agree!

If i were to compare it to relying on SO for your code, then would SO be a better fit because its multiple people. Are models inherently worse than a human community? maybe my question is too generic to answet

Gmail's AI is particularly guilty of robbing the writer of their style.