5 ms·

I appreciate this critique of chat GPT’s writing style since I have often been bothered by it but have struggled to define what was wrong. I think the main feel

I appreciate this critique of chat GPT’s writing style since I have often been bothered by it but have struggled to define what was wrong. I think the main feeling I have is that the words are there to make a sentence, the sentence is there to make a paragraph, whereas any decent human writer uses each word to communicate and entertain. The AI does not have nearly enough nuance in order to make high quality writing. Eventually some AI models may be able to do this but they won’t just be an LLM. Some kind of intelligence needs to exist that attempts to understand underlying fundamentals rather than just finding the most likely next word.

bartvk 2y ago
Why would you struggle to define what's wrong? It's overly verbose and weirdly flowery. I wish LLMs were better. But currently, when I read that drivel, I feel repulsed.
- bqmjjx0kac 2y ago
  It's just a new flavor of uncanny valley IMO, but this time it's an approximation of communication/language. I am in awe of the technological breakthroughs that brought us this far, but disappointed in how it's being intentionally overhyped to devalue human experts like writers and software engineers.
- SatvikBeri 2y ago
  There's a lot of very verbose and flowery writing that's much better than ChatGPT, e.g. anything by Dickens.
  - ZiiS 2y ago
    LLM's target average; 50% of human writing is above average, not just one in a generational talents.
- grues-dinner 2y ago
  I think they write like this partly because most of a generation or more of bad "copywriters" have been educated by teachers, other bad/amateurish writing, SEO BS artists, and even their own clients that this is good writing. Billions of words of this dreck fills the Internet. Everything is an amorphous "experience", is "indulgent" and "unforgettable" and comes in "bespoke packages". Everyone is generically "committed" and "passionate" to their "dream". Read the blurb on any packaging and it's the same empty, flowery pap. It sounds good on a surface level, but it's empty linguistic calories. It's Instagram writing. Lord Dorwin would write like that. It's also tricky to prompt an LLM as you would an (engaged) human copywriter, because the human has extra cues like "why am I being asked to do this anyway", "who asked me" and "what about this coffeeshop is special but isn't in or isn't emphasised by the prompt/original copy". Plus extra research they may do. By the time you've prompted it carefully enough, and verified the emphasis is where you want it, you're basically just writing it. Of course what the LLM will do here is the same crap job that you can get from a guy of Fiverr that they bang out in a few minutes. Because that's where most copy comes from in the first place and that's the training data. And the LLM and the Fiverr guy have the same lack of investment in the job. And, as the Internet, full of the breathless writing about pasionate, committed experience packages proves, that's good enough for most people. It's a bit like smartphone cameras wrecking the low-end camera market. If you want to sell standalone cameras now, you have to sell very good cameras.
- bryanrasmussen 2y ago
  Overly verbose and weirdly flowery is a style, in some places and times even the default style. However when I read the overly verbose and weirdly flowery text of the late 19th century, as an example, it feels different than the weirdly flowery press release style LLMs favor.
  - grues-dinner 2y ago
    It's a different thing entirely. The former style, in which, I must confide to you, my dear fellow, I have always found an inordinate amount of pleasure in reading, risible though you may reasonably find it to be so, written as it often is in a complex form that my erstwhile English masters would likely consider gauche and lacking in focus or, worse, perhaps even damned as trending to running-on, does rather transport my mind to the place and time of writing in a way that mere graven images, despite their undeniable detail and modern polychrome technologies, do not. Perhaps it's a result that writing and reading were both a mark of education and more complex writing was valued over a quick parse by the reader, which is highly valued today. Or, maybe, the use of hand- or type-writing led to train-of-thought subclause-rich writing, because it's hard to recast the sentence without a cursor, and one iteration of fair copying was probably enough for most purposes. Furthermore, much less writing was even done, far fewer people did it and it was expensive and laborious to do, transmit and reproduce. And what was written was often not written by, or intended for reading by, people without expensive educations. Most people probably read a lot more text now than most people did back then, and the do it faster and more disposably. Things these days are often written to be quickly scanned, perhaps only by automatic search crawlers, have keywords recognised and pulled out then the chaff is discarded. Which works better with short, simple sentences. And prominent keywords are important, leading to self-reinforcement of both a set of shibboleth words that "all good copy" should have and how and when those special words are to be used.
  - bartvk 2y ago
    Exactly. Upon smelling the LLM excretion, I immediately get this feeling that I was tricked and I get a bit pissed off.
- Zondartul 2y ago
  My hunch is that since LLMs are trained on a per word basis (okay, per-token), vacuus verbosity is overrepresented. If you have one normal sentence and one overly verbose, the latter will have more tokens and therefore more weight.
andy_xor_andrew 2y ago
the style of ChatGPT always reminds me of bad fanfiction. Which I guess makes sense. There is a lot of bad fanfiction floating around on the web. The hallmarks of this style are easy to spot. Lots of vacuous sentences, which (usually) make grammatical sense but no logical sense. It's very "first-draft-esque". Lots of sentences which, if read a second time, signify nothing, or otherwise just don't hold up to scrutiny.
- cratermoon 2y ago
  The term of art in the writing/editing business is "throat-clearing". A lot of words that don't say much before the action really starts.
HeatrayEnjoyer 2y ago
Everyone misses the fact that ChatGPT's style is intentional, to make it obvious that ChatGPT wrote it. Before RLHF they don't sound like that at all.
- OrigamiPastrami 2y ago
  Is there an LLM I can interact with right now that doesn't sound like a chat bot? I'm very interested in a chat bot that sounds like a person - hell I'd pay money for it.
  - stavros 2y ago
    There are thousands of LLMs, tuned for different purposes. You could also probably prompt GPT in the right way and achieve that.
    - OrigamiPastrami 2y ago
      No, I cannot prompt ChatGPT and get a response that sounds human. You may think that, but in my experience ChatGPT is fucking awful and I was asking if there an LLM that doesn't sound obnoxious. Obviously this is all subjective, but OP implied it would be easier to make an LLM that doesn't sound like a stochastic parrot.
      - stavros 2y ago
        Try Llama 7B, or one of the Mistrals.
        OrigamiPastrami 2y ago
        Thank you for the reply - I'll give them a shot.
        stavros 2y ago
        You're welcome! Generally, I've found that the smaller models sound more human, or at least more creative.

bartvk 2y ago

Why would you struggle to define what's wrong? It's overly verbose and weirdly flowery. I wish LLMs were better. But currently, when I read that drivel, I feel repulsed.

bqmjjx0kac 2y ago
It's just a new flavor of uncanny valley IMO, but this time it's an approximation of communication/language. I am in awe of the technological breakthroughs that brought us this far, but disappointed in how it's being intentionally overhyped to devalue human experts like writers and software engineers.
SatvikBeri 2y ago
There's a lot of very verbose and flowery writing that's much better than ChatGPT, e.g. anything by Dickens.
- ZiiS 2y ago
  LLM's target average; 50% of human writing is above average, not just one in a generational talents.
grues-dinner 2y ago
I think they write like this partly because most of a generation or more of bad "copywriters" have been educated by teachers, other bad/amateurish writing, SEO BS artists, and even their own clients that this is good writing. Billions of words of this dreck fills the Internet. Everything is an amorphous "experience", is "indulgent" and "unforgettable" and comes in "bespoke packages". Everyone is generically "committed" and "passionate" to their "dream". Read the blurb on any packaging and it's the same empty, flowery pap. It sounds good on a surface level, but it's empty linguistic calories. It's Instagram writing. Lord Dorwin would write like that. It's also tricky to prompt an LLM as you would an (engaged) human copywriter, because the human has extra cues like "why am I being asked to do this anyway", "who asked me" and "what about this coffeeshop is special but isn't in or isn't emphasised by the prompt/original copy". Plus extra research they may do. By the time you've prompted it carefully enough, and verified the emphasis is where you want it, you're basically just writing it. Of course what the LLM will do here is the same crap job that you can get from a guy of Fiverr that they bang out in a few minutes. Because that's where most copy comes from in the first place and that's the training data. And the LLM and the Fiverr guy have the same lack of investment in the job. And, as the Internet, full of the breathless writing about pasionate, committed experience packages proves, that's good enough for most people. It's a bit like smartphone cameras wrecking the low-end camera market. If you want to sell standalone cameras now, you have to sell very good cameras.
bryanrasmussen 2y ago
Overly verbose and weirdly flowery is a style, in some places and times even the default style. However when I read the overly verbose and weirdly flowery text of the late 19th century, as an example, it feels different than the weirdly flowery press release style LLMs favor.
- grues-dinner 2y ago
  It's a different thing entirely. The former style, in which, I must confide to you, my dear fellow, I have always found an inordinate amount of pleasure in reading, risible though you may reasonably find it to be so, written as it often is in a complex form that my erstwhile English masters would likely consider gauche and lacking in focus or, worse, perhaps even damned as trending to running-on, does rather transport my mind to the place and time of writing in a way that mere graven images, despite their undeniable detail and modern polychrome technologies, do not. Perhaps it's a result that writing and reading were both a mark of education and more complex writing was valued over a quick parse by the reader, which is highly valued today. Or, maybe, the use of hand- or type-writing led to train-of-thought subclause-rich writing, because it's hard to recast the sentence without a cursor, and one iteration of fair copying was probably enough for most purposes. Furthermore, much less writing was even done, far fewer people did it and it was expensive and laborious to do, transmit and reproduce. And what was written was often not written by, or intended for reading by, people without expensive educations. Most people probably read a lot more text now than most people did back then, and the do it faster and more disposably. Things these days are often written to be quickly scanned, perhaps only by automatic search crawlers, have keywords recognised and pulled out then the chaff is discarded. Which works better with short, simple sentences. And prominent keywords are important, leading to self-reinforcement of both a set of shibboleth words that "all good copy" should have and how and when those special words are to be used.
- bartvk 2y ago
  Exactly. Upon smelling the LLM excretion, I immediately get this feeling that I was tricked and I get a bit pissed off.
Zondartul 2y ago
My hunch is that since LLMs are trained on a per word basis (okay, per-token), vacuus verbosity is overrepresented. If you have one normal sentence and one overly verbose, the latter will have more tokens and therefore more weight.

andy_xor_andrew 2y ago

the style of ChatGPT always reminds me of bad fanfiction. Which I guess makes sense. There is a lot of bad fanfiction floating around on the web. The hallmarks of this style are easy to spot. Lots of vacuous sentences, which (usually) make grammatical sense but no logical sense. It's very "first-draft-esque". Lots of sentences which, if read a second time, signify nothing, or otherwise just don't hold up to scrutiny.

cratermoon 2y ago
The term of art in the writing/editing business is "throat-clearing". A lot of words that don't say much before the action really starts.

HeatrayEnjoyer 2y ago

Everyone misses the fact that ChatGPT's style is intentional, to make it obvious that ChatGPT wrote it. Before RLHF they don't sound like that at all.

OrigamiPastrami 2y ago
Is there an LLM I can interact with right now that doesn't sound like a chat bot? I'm very interested in a chat bot that sounds like a person - hell I'd pay money for it.
- stavros 2y ago
  There are thousands of LLMs, tuned for different purposes. You could also probably prompt GPT in the right way and achieve that.
  - OrigamiPastrami 2y ago
    No, I cannot prompt ChatGPT and get a response that sounds human. You may think that, but in my experience ChatGPT is fucking awful and I was asking if there an LLM that doesn't sound obnoxious. Obviously this is all subjective, but OP implied it would be easier to make an LLM that doesn't sound like a stochastic parrot.
    - stavros 2y ago
      Try Llama 7B, or one of the Mistrals.
      - OrigamiPastrami 2y ago
        Thank you for the reply - I'll give them a shot.
        stavros 2y ago
        You're welcome! Generally, I've found that the smaller models sound more human, or at least more creative.