Hacker News Board Hacker News, browsed like a classic message board
Live data from Hacker News

Semantic ablation: Why AI writing is generic and boring

theregister.com

Post reply on HN
Print view View original

231–234 of 234 posts

Re: Semantic ablation: Why AI writing is generic and boring

#231

Post by lucyjojo »

Retric wrote: post #89
svara wrote: post #84

Earlier quoted context omitted.

I think that mostly depends on how good a writer you are. A lot of people aren't, and the AI legitimately writes better. As in, the prose is easier to understand, free of obvious errors or ambiguities. But then, the writing is also never great. I've tried a couple of times to get it to write in the style of a famous author, sometimes pasting in some example text to model the output on, but it never sounds right.

I find most people can write way better than AI, they simply don’t put in the effort. Which is the real issue, we’re flooding channels not designed for such low effort submissions. AI slop is just SPAM in a different context.

you cannot write well if you do not read a lot (you need to develop taste). this disqualifies most people. i included.

Re: Semantic ablation: Why AI writing is generic and boring

#232

Post by lucyjojo »

anon-3988 wrote: post #194
baxtr wrote: post #103

Earlier quoted context omitted.

I think it’s essential to realize that AI is a tool for mainstream tasks like composing a standard email and not for the edges. The edges are where interesting stuff happens. The boring part can be made more efficient. I don’t need to type boring emails, people who can’t articulate well will be elevated. It’s the efficient popularization of the boring stuff. Not much else.

> The edges are where interesting stuff happens. The boring part can be made more efficient. I don’t need to type boring emails, people who can’t articulate well will be elevated. I think that boring emails should not be written. What kind of boring emails do you NEED to be written, but not WANT to write? Those are exactly the kind of email that SHOULD NOT be passed through an LLM. If you need to say yes/no. You don'…

when i pass my writings through ai the output is generally only marginally bigger than the input, and it derisks things a lot making my prose a nice beige.

Re: Semantic ablation: Why AI writing is generic and boring

#233

Post by writingdna »

resiros wrote: post #27
altmanaltman wrote: post #17

Earlier quoted context omitted.

I mean there's tons of better-writing tools that use AI like Grammarly etc. For actual general-purpose LLMs, I don't think there's much incentive in making it write "better" in the artistic sense of the world... if the idea is to make the model good at tasks in general and communicate via language, that language should sound generic and boring. If it's too artistic or poetic or novel-like, the communication would app…

I mean, no one is asking for artistic writing, just not some obvious AI slop. The fact that we all can now easily determine that some text has been written / edited by AI is already an issue. No amount of prompting can help.

The article frames this as "semantic ablation" but the underlying mechanism is more specific: it is distributional averaging. RLHF and DPO reward policies optimize for the modal response given a prompt distribution. That is not a bug in the training process, it is the objective function working as designed. The model learns to produce the response that the median annotator would rate highest, and that response is, almost by definition, the least distinctive one.

What is underappreciated is how much stylistic signal lives in what information retrieval people call "burstiness" -- the tendency for distinctive words to cluster rather than distribute evenly. Hemingway's short declarative stacking, DFW's recursive parentheticals, legal writing's formulaic precision -- these are all bursty patterns that a model trained to maximize expected reward will sand down. You can partially recover it with few-shot prompting, but the model is fighting its own reward gradient the entire time.

The practical question is whether you can encode a style prior that survives the decoding process. The research on authorship attribution (stylometry) suggests the feature set is well-understood -- function word frequencies, sentence length distributions, type-token ratios, syntactic complexity metrics. But nobody has built a production system that uses those features as a constraint during generation rather than just detection.

Post reply on HN