I like the way Ted Chiang puts it:
Some might say that the output of large language models doesn’t look all that different from a human writer’s first draft, but, again, I think this is a superficial resemblance. Your first draft isn’t an unoriginal idea expressed clearly; it’s an original idea expressed poorly, and it is accompanied by your amorphous dissatisfaction, your awareness of the distance between what it says and what you want it to say. That’s what directs you during rewriting, and that’s one of the things lacking when you start with text generated by an A.I.
There’s nothing magical or mystical about writing, but it involves more than placing an existing document on an unreliable photocopier and pressing the Print button.
I think our materialist culture forgets that minds exist. The output from writing something is not just “the thing you wrote”, but also your thoughts about the thing you wrote.
The dialog pushing AI media seems to start from this assumption that I consume media just to have colors and words and sounds enter my face holes. In fact, I consume art and media because I like hearing, seeing, and reading about how other humans experience the same world I do. It’s a form of communication. I like the product but also the process of people trying to capture the bonkers, ineffable experience we all seem to be sharing in ways I would never think of, but can instantly verify.
What’s funny is, due to the nature of media, it’s kind of impossible to not communicate something, even if the artwork itself is empty. When I see AI media I see the communication of a mind that doesn’t know or give a shit about any of this. So in their attempt make filler they are in fact making art about how inarticulate they are. It’s unintentional, corporate dadaism.
Yes, this is it right here. The whole point of art is communication and connection with another human being.
Not even necessarily a human being! I’d appreciate the fuck out of art if any species made it. But there must be more than uncaring, unfeeling, probabilistic interpretation of input data.
Hell yeah, otherkin inclusive language
The people pushing AI don’t like like hearing, seeing, and reading about how other humans experience the world. They actually do just want flashing colors and sounds poured into their face holes. They’re basically incapable of understanding art.
I think the reason so many AI bros are conservative is that conservatives have historically had really bad taste in art/media, so they see the drivel AI creates and think, “oh wow, it looks just like what the artists make,” not realizing that they don’t have the eye to see what it’s missing.
Tons of shit games are going to have lots of dialogue written by AI. It’s very likely that those games would have had shit dialogue anyway.
Sure, but human-written shit still had that human touch. It could be unintentionally funny, it could be a mixed bag that reaches unexpected heights at times. AI writing is just the bland kind of bad, not the interesting kind of bad.
Great point. There’s no opportunity for “so bad it’s good”. The Room wouldn’t have been a thing if Tommy used AI.
You’re absolutely right, Lisa!
deleted by creator
Writer having toyed with AI, here : yeah, AI writing sucks. It is consensual and bland, never goes into unexpected territory, or completely fails to understand human nature.
So, we’d better stop calling AI “intelligence”. It’s text-prediction machine learning on steroïds, nothing more, and the fact that we’re still calling that “intelligence” says how gullible we all are.
It’s just another speculative bubble from the tech bros, as cryptos were, except this time the tech bros have made their nazi coming out.
AI is just a marketing term, there’s nothing intelligent about it. Its simply Large Language Models, databases that predict what should go next. Its like asking the prediction bar when you are typing to write a story.
You can make it stylized dialogue but it’s just surface mannerisms. Underneath it’s still the same bland AI
That means in about 6 months or so the AI content quality will be about an 8/10. The processors spread machine “learning” incredibly fast. Some might even say exponentially fast. Pretty soon it’ll be like that old song “If you wonder why your letters never get a reply, when you tell me that you love me, I want to see you write it”. “Letters” is an old version of one-on-one tweeting, but with no character limit.
My hot take. It will never get to even a 6/10. I bet it will just spit out 3/10 faster and faster, most likely.
Only if you assume that its performance will continue improving for a good while and (at least) linearly. The companies are really struggling to give their models more compute or more training data now and frankly it doesn’t seem like there have been any big strides for a while
Yeah… Linear increases in performance appear to require exponentially more data, hardware, and energy.
Meanwhile, the big companies are passing around the same $100bn IOU, amortizing GPUs on 6-year schedules but burning them out in months, using those same GPUs as collateral on massive loans, and spending based on an ever-accelerating number of data centers which are not guaranteed to get built or receive sufficient power.
What improvements have there been in the previous 6 months? From what I’ve seen the AI is still spewing the same 3/10 slop it has since 2021, with maybe one or two improvements bringing it up from 2/10. I’ve heard several people say some newer/bigger models actually got worse at certain tasks, and clean training data is pretty much dried up to even train more models.
I just don’t see any world where scaling up the compute and power usage is going to suddenly improve the quality orders of magnitude. By design LLMs are programmed to output the most statistically likely response, but almost by definition is going to be the most average, bland response possible.
I doubt that. A lot of the poor writing quality comes down to choice. All the most powerful models are inherently trained to be bland, seek harmony with the user, and generally come across as kind of slimy in a typically corporate sort of way. This bleeds into the writing style pretty heavily.
A model trained specifically for creative writing without such a focus would probably do better. We’ll see.
sadly, a bad first draft in five minutes usually beats a thoughtful solution in three days
edit: oops i thought we were talking about code. nvm, people actually look at prose.
