mercer
I've had some interesting conversations about the pre-AI slop and how the distinction is often not too meaningful. I mean, on some level /knowing/ that the corporate slop was written by a person kind of loses its meaning when one considers the amount of filtering, rules, rewriting, and so on is involved by a string of people who either don't care about what they write or actively dislike it.
A lot of this stuff often had at least flickers of a human soul behind it. I imagine for a lot of the soaps, hallmark movies, and romance novels this might've made some difference, even if just subconsciously recognizing the author(s) and building up some kind of image around the other by yourself or with others.
I'm reminded of some of the truly awful 'worship' songs I grew up with, and how some of the authors (often of course sticking close to the source material) even had a kind of following. some of these songs were just a little 'too' predictable, but others felt pretty much like they were written by AI, except back then none of us imagined that AI could really do this back then.
Or the pastor I spoke to who was convinced that chatgpt had access to his more personal notes, because as an experiment he made it write a sermon, and it (probably? obviously?) extrapolated from the prompt the more specific theology he followed/ascribed to.
I'm not knocking the value of a sermon or song, even if boring and predictable, or perhaps even when AI-written. they often explicitly /don't/ serve the purpose of being novel or new, but I find it super interesting how these aspects of real life are, to a degree, not too hard to replace with AI.
Thankfully for this pastor, most of his value involved human connection, knowing whoever just married, died, got sick, and having his full humanity on display from the pulpit!
I'm not saying that we didn't have anything like that before tv, or that specific individuals or groups throughout history might not have had something similar, but I do feel TV, and especially its audio-visual nature, really changed something in a way that, say, the printing press never quite did.
EDIT: and to add, my feeling on how many people seem to use LLM's is that in a way it's extra insidious because it's /tailored/, often 'puerile' interaction.
> JOSE: Yeah, so what happened is that it was the old concurrency story in which the Clojure audience is going to be really, really familiar. I’ve learned a lot also from Clojure because, at the time I was thinking about Elixir, Clojure was already around. I like to say it’s one of the top three influences in Elixir, but anyway it tells this whole story about concurrency, right?
Anecdotally what I'm hearing is that this is pretty much how LLMs are helping programmers get more done, including the work being less enjoyable because it involves more verification and rubber-stamping.
For the business owner, it doesn't matter that the nature of the work has changed, as long as that one person can get more work done. Even worse, the business owner probably doesn't care as much about the quality of the resulting work, as long as it works.
I'm reminded of how much of my work has involved implementing solutions that took less careful thought, where even when I outlined the drawbacks, the owner wanted it done the quick way. And if the problems arose, often quite a bit later, it was as if they hadn't made that initial decision in the first place.
For my personal tinkering, I've all but defaulted to the LLMs returning suggested actions at logical points in the workflow, leaving me to confirm or cancel whatever it came up with. this definitely still makes the process faster, just not as magically automatic.
A lot of the issues I'd have when 'pretending' to have a conversation are much less so when I either keep things to a single Q/A pairing, or at the very least heavily edit/prune the conversation history. Based on my understanding of LLM's, this seems to make sense even for the models that are trained for conversational interfaces.
so, for example, an exchange with multiple messages, where at the end I ask the LLM to double-check the conversation and correct 'hallucinations', is less optimal than something like asking for a thorough summary at the end, and then feeding that into a new prompt/conversation, as the repetition of these falsities, or 'building' on them with subsequent messages, is more likely to make them a stronger 'presence' and as a result perhaps affect the corrections.
I haven't tested any of this thoroughly, but at least with code I've definitely noticed how a wrong piece of code can 'infect' the conversation.