I think you're right that a big thing missing is for an AI to understand what might be interesting to a human. It doesn't really have that. But I think it does have everything else. You mention that AIs can't write anything creative - all it does is combine things its been trained on potentially with a randomness factor. But I'd argue that's basically exactly what humans do. They have their experience to draw from and they combine novel things together because of their novel set of experiences. The differences is that (some) humans know which random ideas they come up with are interesting, and which to throw out. If one could train an AI to recognize interesting things, that would be a very valuable module.
If your child asks for a bedtime story, you have to improvise on the spot. Which is what LLMs do. Whereas a human author makes plans. A human makes outlines to enforce coherency. (Indeed, the Game of Thrones extension did use multistage outlining, but the github has been deleted, so I don't know the details of the strategy.) A human knows which steps require creativity. But an LLM can be instructed to do these steps. Something like Agent-GPT can probably generate these steps.
If you start with the plan to write a story about explaining plumbing to a child and ask it to add a detail to mix up the story, what kind of details does it suggest? Is the motorcycle theme so implausible? Many people claim that LLMs are valuable specifically for brainstorming. But those humans do the filtering. Can the model filter good ideas? Why not? One way to filter is to just repeat back the ideas and ask which are better. Another way is to ask it to flesh out each idea or finish each story and then ask it which produced the best story, or even specifically which meshed with the theme. Maybe the child can recognize good stories without analysis, but the LLM can perform analysis.
There's a long history of people claiming that LLMs can't do things that they can, if you just ask them, or maybe if you ask them right. I don't know that they can do this, but this very theoretical argument seems unconvincing to me. Is human creativity so special? I don't know, but I doubt it. Can LLMs judge good stories? Sounds hard. Now that you've planted your flag, test it. What would it tell you if you are wrong?
Another thing humans do is drafts. Would this help an LLM, or does it produce the same quality of writing each time? Why do human drafts improve? One possibility is that humans alternate between paying attention to different structures in different drafts, eg, alternating between improving sentences and improving paragraphs. If the LLM has only one level of attention, this wouldn't help, although it could alternate between temperatures. Another possibility is that humans have to take a break to reset the grooves in their brains, their random number generators so that they can find more options. LLMs can't. A possibility that I mentioned above as possibly applicable to LLMs is that some ideas have to be expanded before they can be evaluated.
I've been playing roleplaying games for fifty years. In that time, I've tried a lot of systems for generating things by random dice rolls: dungeons, superheroes, outer space adventurers, distant planets, solar systems, alien races . . . That has the kind of randomness you refer to, but it's very poor at utility; large numbers of the things I generate, in whatever category, are just really incoherent and uninteresting. And it doesn't seem possible to avoid this. If I want something that makes for a good story or a good scenario, I need to generate a ridiculous number of entities and select from them, or I need to come up with my own original ideas. That seems as if it may be kind of like the issues you're talking about.
Yes, I've never used the random tables in RPG sourcebooks to generate character backstory or traits, because of just that. I'll read them over for ideas, but I want to choose myself what sort of character will click with me.
Perhaps drafting and looking at the draft from a new direction would help LLM's. I've heard that when an LLM outputs some source code that might have security flaws, if you then feed it the same code and ask it to check for security flaws, it'll often find the result. If you feed it a story idea or paragraph it generated and ask it to evaluate it...
Well, it depends on whether the LLM can recognize goodness in a story. In theory, it could to some extent thanks to all its training data. But it would be cribbing off that.
I think you're right that a big thing missing is for an AI to understand what might be interesting to a human. It doesn't really have that. But I think it does have everything else. You mention that AIs can't write anything creative - all it does is combine things its been trained on potentially with a randomness factor. But I'd argue that's basically exactly what humans do. They have their experience to draw from and they combine novel things together because of their novel set of experiences. The differences is that (some) humans know which random ideas they come up with are interesting, and which to throw out. If one could train an AI to recognize interesting things, that would be a very valuable module.
If your child asks for a bedtime story, you have to improvise on the spot. Which is what LLMs do. Whereas a human author makes plans. A human makes outlines to enforce coherency. (Indeed, the Game of Thrones extension did use multistage outlining, but the github has been deleted, so I don't know the details of the strategy.) A human knows which steps require creativity. But an LLM can be instructed to do these steps. Something like Agent-GPT can probably generate these steps.
If you start with the plan to write a story about explaining plumbing to a child and ask it to add a detail to mix up the story, what kind of details does it suggest? Is the motorcycle theme so implausible? Many people claim that LLMs are valuable specifically for brainstorming. But those humans do the filtering. Can the model filter good ideas? Why not? One way to filter is to just repeat back the ideas and ask which are better. Another way is to ask it to flesh out each idea or finish each story and then ask it which produced the best story, or even specifically which meshed with the theme. Maybe the child can recognize good stories without analysis, but the LLM can perform analysis.
There's a long history of people claiming that LLMs can't do things that they can, if you just ask them, or maybe if you ask them right. I don't know that they can do this, but this very theoretical argument seems unconvincing to me. Is human creativity so special? I don't know, but I doubt it. Can LLMs judge good stories? Sounds hard. Now that you've planted your flag, test it. What would it tell you if you are wrong?
Another thing humans do is drafts. Would this help an LLM, or does it produce the same quality of writing each time? Why do human drafts improve? One possibility is that humans alternate between paying attention to different structures in different drafts, eg, alternating between improving sentences and improving paragraphs. If the LLM has only one level of attention, this wouldn't help, although it could alternate between temperatures. Another possibility is that humans have to take a break to reset the grooves in their brains, their random number generators so that they can find more options. LLMs can't. A possibility that I mentioned above as possibly applicable to LLMs is that some ideas have to be expanded before they can be evaluated.
I've been playing roleplaying games for fifty years. In that time, I've tried a lot of systems for generating things by random dice rolls: dungeons, superheroes, outer space adventurers, distant planets, solar systems, alien races . . . That has the kind of randomness you refer to, but it's very poor at utility; large numbers of the things I generate, in whatever category, are just really incoherent and uninteresting. And it doesn't seem possible to avoid this. If I want something that makes for a good story or a good scenario, I need to generate a ridiculous number of entities and select from them, or I need to come up with my own original ideas. That seems as if it may be kind of like the issues you're talking about.
Yes, I've never used the random tables in RPG sourcebooks to generate character backstory or traits, because of just that. I'll read them over for ideas, but I want to choose myself what sort of character will click with me.
Perhaps drafting and looking at the draft from a new direction would help LLM's. I've heard that when an LLM outputs some source code that might have security flaws, if you then feed it the same code and ask it to check for security flaws, it'll often find the result. If you feed it a story idea or paragraph it generated and ask it to evaluate it...
Well, it depends on whether the LLM can recognize goodness in a story. In theory, it could to some extent thanks to all its training data. But it would be cribbing off that.