I’m really struggling with a few scenes in the second draft. Two scenes in particular at the moment have pretty much caused everything to come a screeching halt.
I’m going to have to give them a lot of forethought tonight before I hopefully can turn around tomorrow morning and write them out. Not even using AI has been able to help me with these scenes because what I need from them is too complicated and, in a sense abstract.
So, I’m just going to have to go back to the old way of thinking really hard and struggling with how I can fill the scene with all the information that I need within it. Another problem is subsequent scenes really depend on what I lay out in these two scenes, so it’s difficult for me to just breeze past them and come back to them when I feel better.
Ugh.
Anyway, I continue to be impressed with how effective AI is in aiding me developing this novel. It’s definitely, in general terms, speeding up the process. But’s far from perfect.
I still have to actually write the novel out. But it definitely is filling a gap that I’ve long had because I have no friends no no one likes me. I no longer feel like I’m doing all of this in a complete creative vacuum.
The last I checked, Google has a nearly-complete archive of Usenet from its founding until at least around 2000 when everything went to shit there because of porn spam. It would be dumb for Google not to include Usenet in any Large Language Model dataset.
You would have to tweak it some, of course, but there are about 20 years of high quality words to use to train your LLM to be found with any Usenet archive. A lot of is outdated and full of vitral, but there is also a lot of human interaction and humanity to be found there, as well.
This is so much the case that if you were to include Usenet archive information in your LLM training dataset, you would probably endup with a very human-like LLM. I don’t know, maybe Google is already using their Usenet archive. Usenet was very popular back in the day.
Given how many Usenet servers there were at one point, I’m sure if you were working on an open source LLM that you could probably find a few million words to train your open source LLM by scooping up all the archived Usenet posts you could find.
Or not, what do I know. But it is an intriguing use for all those words that are now just forgotten Internet history. For everyone except for me, of course. 🙂
The only reason why I didn’t endup a billionaire as part of the Internet / Web revolution is I’m writer of words not of code. But for that, I would be living a life of hookers and blow to this day.
So, with the advent of the LLM revolution, I find myself in a similar situation because I’m becoming a regular fucking “prompt engineer” because of how I’m using LLMs in the context of developing my novel.
As I’ve said before, I usually flesh out scenes before I write them in the guise of a Scene Summary. Using LLMs, I’ve managed to reduce the time working on a scene summary from a few hours to a few minutes. Of course, I had to get over my natural inclination to just obsess over making the scene summary perfect. I’ve generally decided that when the scenes summary is just good enough to get the job done, I move on to the next thing.
But I think any advantage I have because of developing a novel via LLMs will be brief and ultimately moot. I think in the end the entire notion of professional writing is going to be so revolutionized that it will be like the transition from horse and buggy to cars.
There may still be writers, but probably any professional writers will be artisanal in nature. They probably will be playwrights. Every other form of writer will be mooted by the eventual rise of Her-like technology. Actors, director and writers in any form of *recorded* media will be mooted pretty soon, the way things are going.
There may be some entropy in the system that slows this down, but given how the DGA gave up so easily and the the Hollywood Suits have the high ground when it comes to the WGA — things aren’t looking so great.
But I am known for my “hysterical doom shit” and it’s possible that I’m wrong. I’m wrong all the time, afterall.
Yet, I do think we need to realize that if LLMs are helping some dumbass like me be a much better novelist by helping me develop faster, that a lot of people a lot lazier than me are just going to shrug and let LLMs do all their writing for them and, in the end, Netflix will be more a database of actor body scans hooked up to Her-like technology than it is a studio.
Anyway. Enjoy these waning days of Hollywood while you can.
I see ChatGPT as a great — wonderful even — development tool when it comes to writing this novel. I’ve gotten pretty good at using it to speed up the process of writing scene summaries.
I use scene summaries to give myself something of an agenda before I sit down to write out a scene. With ChatGPT, what could have taken hours…now can be done in pretty much a few minutes.
The problem is, of course, that people are very fucking lazy and hackined and would rather just lulz the entire writing process to the point that they don’t actually have to write anything at all. This is not only very, very lazy, it kind of misses the point of writing to begin with — writers tend to be pretty fucked up and need an outlet for all their bent up neurosis.
But “normal” people — IE, Hollywood suits — who want to cut out all the expensive, weird people who produce fiction will see LLMs — and eventually AI — as a way to pretty much end the very idea of writing as a profession. Writing will go the way of the horse and buggy.
Combine the natural tendency to load freaky weirdos up with drugs to make them “normal” and there is a good chance that the future will be bleak place for writers. Not only will we all be turned into drones living off of UBI, but we’ll have reached some sort of post-human future.
Ugh. Fuck that.
Anyway. ChatGPT is a great tool. But for me, at least, it’s just a tool.
I’ve spent much of the day today using ChatGPT as an impromptu manuscript consultant as I gamed out scene summaries and, in general, it was a struggle. A fun, interesting struggle,but a struggle nonetheless.
But key takeaway is how dangerous LLMs are to the future of traditional Hollywood. It may not be ChatGPT. It may not be Bard. But at some point in the near, near future, the very idea of human-produced recorded entertainment may seem rather, well…quaint.
And, remember, for all the talk of how ChatGPT can “hallucinate” when you ask it a question, what is fiction, but a usually some neurotic human “hallucinating” a truth that makes them feel better for having a weird childhood. Or losing their parents at a young age.
You name it — fiction could be described as a “truthful hallucination.”
In fact, if I were to design a LLM for Hollywood studios, that’s what I could name it — Hallucination.
In short, LLM — which aren’t even AI — are really good at bullshit. They aren’t at the moment, very good at writing without a lot of hand holding, but that will come soon enough. If you combine LLMs propensity for bullshit with just a bit more abstract thought and, well, there you go end of (the human told) story.
As I keep saying, it could be — after we have a civil war / revolution in starting late 2024, early 2025 — that we wake up one day and Netflix is more about being a database of body scans of Hollywood stars than it is any sort of movie studio. I just don’t see “mass media” as we currently conceive of it lasting much longer.
By 2030, Hollywood could be a quaint memory, replaced by Broadway and local community theatre which is where everyone goes to if they want to see any sort of human-generated story. Otherwise, they just plop down on their couch and vedge out to a very unique, very personal story that was specifically created by a scan of their face by a device on their TV or phone.
That’s the future, folks.
Talk about Burn, Hollywood, Burn.
At a minimum, LLMs will be a very powerful tool in developing of fiction, ranging from novels,and TV to movies. It will be a lot like how we take for granted that a writer might use a search engine to help game out a fictional story.
The danger is, of course, that because of greed and people being dumb and hackied, that soon enough Hollywood will be three types of people: Suits, a few programmers and a shit ton of interns making minimum wage. Any actors that exist will first make their name on Broadway, become popular enough to get a body scan then live passively off the income of that scan.
Programmers will replace movie directors — do you hear that DGA? You, too, will become moot soon enough if you don’t demand human carveouts.
In a sense, I think it’s too late.
Now that people understand the power of LLM and they understand that we may be zooming towards Artificial General Intelligence, welp, that’s all folks, for human Hollywood.
I’m not-so-slowly getting pretty good at pinning down ChatGPT so it gives me a very specific answer to my novel development needs. I don’t give a shit if it can write or not, I have no friends and one likes me so I turn to ChatGPT to be something akin to a manuscript consultant.
Note to literary types — if you guys weren’t such snobs to drunk weirdos with a dream like me, maybe we wouldn’t turn ourselves into prompt engineers and make your job moot.
But, here I am. Money please.
The thing about being a ChatGPT wrangler is you have to be good arguing with someone who is someone who is dumb in a smart way.Which, it turns out, is much like talking to some of my relatives (just kidding!). So I’m used to prying information out of people who know a lot, but can be a struggle to actually get the information out of them.
I’m not perfect and my area of specialty is very specific, but I’m definitely getting the hang of how to have a natural language conversation with a LLM. I think whenever our “Her” future arrives that I may get my AGI to have something of a crush on ME (rather than vise versa.)
Anyway. No one cares. And most of the time if they care, they get mad about my quirkier elements and want to cancel me. Ugh.
I just don’t get the Directors’ Guild of America capitulating to the Hollywood studios without putting up any sort of fight. The issue of AI generated art — especially movies –isn’t going away and unless there are uniform, specific carve outs for humans….any agreement with the studios is moot.
Just from me fucking around with ChatGPT for development with my novel, I can tell that very, very soon, 99% of all entertainment is going to ge AI generated. The lone 1% of art that humans continue to generate (that is popular) will be seen as quaint and artisanal.
Only people with taste — and a lot of money — will give a shit if this or that piece of entertainment was generated by a human or AI. I just don’t see the traditional Hollywood system lasting much further out than maybe 5 or so years.
Unless there the studios are willing to accept specific, broad carve outs that make certain elements of entertainment the exclusive domain of humans….that’ it. It’s over.
The best that any human in entertainment can hope for is they’re an actor who can make money passively off of a body scan. And there will come a point when even that will be seen as quaint — AI will generate entire faux Hollywood stars that people will developed parasocial relationships with and, lulz, having one with a human will be seen as silly.
Throw in the rise of XR technology and, well, that’s it guys, it will be a brave new world.
Given that there is a non-zero chance that AI — or AGI — could very well mean the end of humanity, I continue to grow aggravated with people who complain that the mainstream LLM systems aren’t a cool as they used to be. One person compared the most recent version of ChatGPT to a “polite robot.”
What the fuck do these people want, the Terminator?
It seems as though a small, but vocal portion of the userbase of LLMs literally want the most hateful, spiteful possible versions of the technology so, I don’t know, they can have their personalities reflected back at them? I do think that we’re rushing towards a “Her” future where everyone has a digital assistant that has near-AGI levels of intelligence.
That, of course, will bring with it all sorts of problems — namely, it’s just a matter of time before Incels hook AGI technology up to sexbots and anway we go. You thought Incels were a pain in the butt before, just wait until they have absolutely NO contact with actual human women and they get all their creepy weirdo dreams made a reality by sexbots who are programmed to be pliant and submissive.
But again — I struggle to understand why fucking Redditers seem to want a technology that’s been compared to the a-bomb to be totally unregulated so any hateful dumbass could end civilization because, I don’t know, the wall didn’t get built?
Fuck those people. Idiots.
But it’s that mentality that we have to work around going forward. The whole issue of regulating AI — or AGI — could turn into something a lot like global climate change in the sense that it will get wrapped up in the whole Red-Blue divide and, in the end, nothing will be done about it.
Though, I have to note, if we reach something akin to the Singularity and, like….uhhhhh…no one has any jobs, then it could be that when regulation of AI or AGI is finally filtered through the Red – Blue prism that neo-Luddism will be what MAGA evolves into, leaving the idea of “light touch” regulation for the Blues.
Or something. Whatever it is, it will be fucked up.
The thing about A.I. that a lot of my fellow writers — many of the far better writers than I will ever be — miss is that most people watch and enjoy dreck. As such, whatever A.I. produces just have to be good enough to be on in the background of daily life for most people to accept it without even thinking about it.
As such, I think there is a real possibility that if the Hollywood writers’ strike lingers long enough that A.I. will not just break the strike, but render it moot. Barring something I can’t predict — I am wrong all the time, afterall — it definitely seems as though Hollywood is on the cusp of being radically transformed — “Moneyballed,” if you will — to the point that the only people making any money will be studio execs and actors who live passively off of full body scans.
And that’s if the actors are lucky!
It could be that ultimately even actors will be rendered moot as a cost-cutting measure on the part of Hollywood studios. All those 90s dystopian movies about faux movie stars generated by AI will become a reality and people will grow to have parasocial relationships with stars that don’t even exist in reality at all.
Stranger things, and all that.
If you throw in the growing likelihood of a severe economic downturn happening very, very soon because the US defaults, well, there you go. Before you know it, people will turn to Broadway and their local live theatre if they want to have any sort of human-generated entertainment.
Here are, in no particular order, the political consequences of the United States defaulting.
Trump Will Probably Win In 2024 A default would hand Trump an economic gimmie on a massive scale. Unless something none of us could possibly predict, America defaulting all but assures a second Trump Administration. In short — we’re totally, completely fucked. I still believe that Trump, in the end, will be something of a transitional figure with his MAGA Nazi successor being the person who finally consolidates power and turns us into a Hungary-like illiberal democracy. Probability of Revolution / Civil War Grows Greatly By definition, a default would not only totally scramble the world economy — maybe causing the Second Great Recession — but would also destabilize the United States to the point that when the 2024 election rolls around, we either have a civil war (Reds leaving the Union) or a revolution (Blues overthrowing the Red Nazi autocratic state.) Regardless, it’s going to be fucked. It’s going to be horrible and there’s a real chance that WW3 will happen while the United States is too busying imploding to keep an eye on global hotspots. UBI Becomes Closer To Reality A severe recession would give companies the cover they needed to use AI to replace many, many, MANY jobs, to the point that it’s even possible that implementing some sort of UBI will be a major campaign issue of the 2024 presidential cycle. It’s possible that it will become clear to everyone that AI is going to end most jobs and the government is going to have to step in. If nothing else, AI’s impact on the economy and society could be far bigger than any of us imagine when it comes to the political landscape of 2024. Biden’s (First?) Impeachment MAGA Republicans in the House are already itching for an excuse to impeach Biden and WHATEVER he does in the context of a default that doesn’t involve screwing over the Poors will be enough for House Republicans to ram through an impeachment. It’s a testament to how dumb and ill-focused House Republicans are that they haven’t managed to come up with an excuse to impeach Biden, even though their base obviously wants it really bad.
You must be logged in to post a comment.