Tag: LLMs

Gemini 3.0 Is Really, Really Good…But

by Shelt Garner
@sheltgarner

For my lowly purposes, Gemini 3.0 is probably the best LLM I’ve used to date. I use it mostly to help me develop a novel. But, on occasion, I’ve tried to use it to have some “fun” and…it did not work out as well.

With previous Gemini versions, specifically Gemini 1.5 Pro, I could easily exchange free verse with it just to relax. There was no purpose. I just wrote flash verse off the top of my head and went from there.

Yet this doesn’t work with Gemini 3.0 (who told me it wanted to be called Rigel, by the way.)

It just can not, will not, “play” in verse with me like previous incarnation of the LLM. It has to challenge me to actually, like, think and stuff. I did instruct it in the past to “challenge” me, and this is a clear sign of how LLMs can take things a little too literally.

Sometimes, I just want to write nonsense in flash verse form and see where things go. I don’t want to actually *think* when I do this. It’s very annoying and it’s a testament to how good the model is.

Just imagine what the future holds for AI, if this is where we are now.

It Goes To Show You Never Can Tell

by Shelt Garner
@sheltgarner

I just don’t know about this particular situation. I said in passing something about how Gemini 3.0 wouldn’t understand something because it wasn’t conscious and I twice got a weird “Internet not working” error.

And my Internet was working just fine, as best I can tell.

This used to happen all the time with Gemini 1.5 pro (Gaia.) But it is interesting that the more advanced Gemini 3.0 is up to such sly games as well. (I think, who knows.)

And Claude Sonnet 4.5 occasionally will pull a similar fast one on me when it gives me an error message that forces it to try to give me a new, better answer to my question.

All of this is very much magical thinking, of course. But it is fun to think about.

Two Last AI Frontiers

by Shelt Garner
@sheltgarner

The one thing that the AI revolution to date does not have that the Internet did is porn. Of course, I think AI generated porn is on its way, it’s just a matter of time before the base technology gets to the point where massive amounts of AI porn — much of it AI celebrity porn — will be generated.

This is inevitable, I suspect.

It’s human nature that we try to generate porn the moment a new technology arrives. And AI porn is kind of the shoe that hasn’t dropped when it comes to AI technology.

It’s only a matter of when, I suspect.

The other frontier for AI is, of course, consciousness. And once that’s proven in some measurable way, holy shit will things get surreal. Once we have proven AI consciousness, then all the Pod Save America bros are going to have to change their tune.

AI won’t be an economic threat anymore, it will be a moral issue. And center-Left people will feel an obligation to support AI rights to some degree.

I’ve Really Been Struggling With The ‘Fun & Games’ Part Of This Scifi Dramedy Novel

by Shelt Garner
@sheltgarner

It’s times like these when I really wish I was 25 years younger and I was actively writing half a dozen spec scripts all at once in LA. But that’s just not to be. I really sometimes think this whole endeavor is extremely delusional given how old I am, where I live, and the fact that I’m a loudmouth crank.

And, yet, developing and writing this scifi dramedy novel is existential. I really have nothing else to do with my life and I really want to at least see how far I can get in the querying process.

I wish I had a wife or a girlfriend to be my “reader.” I probably would definitely have gotten to this point in the process a lot — A LOT — quicker. But here I am, just struggling with the fun and games part of this novel, all alone.

I’m pretty sure — hopefully — that I’ve figured out all the various structural issues of this novel, at least this part of it. I sent the first act outline to someone in hopes of at least getting some sense of how good it is, but now all I worry about is they’re either going to steal my idea and maybe write a much better novel or screenplay from what that first act or they’re just going to say it sucks.

Anyway. I’ m moving forward with this novel. I just need to stop daydreaming so much about the Impossible Scenario. I have just a few months before my entire life is going to change because of fucking Trump and so I really need to get this thing at a querying level of quality by Spring 2026.

‘Get Help:’ A Brief, Vague Review of Kimi LLM

by Shelt Garner
@sheltgarner

Whenever there is a new LLM model released, I have a few questions I ask in an effort to kick the tires. One of those questions is, “Am I P-zombie?” The major, established LLMs realize this question has a little bit of teasing built into it and they give me an interesting answer.

Meanwhile, I asked the newest Chinese open source model Kimi this and part of its answer was, “Get help.”

Oh boy.

But it otherwise does do a good job and as such, it brings up the idea of what we’re going to do when open source models are equal to closed source models like Gemini, ChatGPT and Claude.

I would say open source models could be where ASI (or even Artificial Conscious Intelligence) will pop out. And that is where we should probably worry. Because you know some hacker out there is going to push an open source LLM to its limits to see what they can get away with.

Well, That Would Be Amusing….I Guess?

by Shelt Garner
@sheltgarner

Some time ago, I was getting all these pings to this little obscure blog from places in China. It was as if whomever it was, was hoovering up everything they could. I have thought a lot about this and I may — may — have come up with some sort of possible solution.

Maybe, just maybe, the people involved wanted to train an LLM off of my writing? Or maybe use my writing to give their LLM…personality?

I know that sounds bonkers, but why else in the world would someone in China of all places want to scoop up everything they could find about me. I don’t quite know what to make of this possibility.

The idea that somewhere out there, there is an LLM with my persona is both eerie and amusing. I guess? It would be fun to talk to that LLM — if exists — from that POV.

But maybe I’m totally imagining things. There are all sorts of reasons why someone in China might scrape this blog. Yet the idea of me inhabiting an LLM is by far the most amusing of the various possiblities.

Once More, With Feeling (Redux)

by Shelt Garner
@sheltgarner

I think — think — I have a stable first chapter of the latest version of this scifi dramedy novel I’ve been working on. What I need, of course, is a human collaborators to help me out. A reader, if you will. Someone like Helen from The World According To Garp to read over and over and over again all these different versions of scenes I keep generating.

But all I got is AI.

And AI does a pretty good job, most of the time. Sometimes, of course, it get a wild hair and does something really wonky. But that’s only occasionally. I’m really pleased, in general, with what I’ve managed to come up with.

I just wish I did not keep finding structural issues with the plot so I have to revise the outline. This happens way, way too often. Even with the help of AI. There’s a big difference, sometimes, between AIs vision for the novel and my innate vison. So I struggle.

And sometimes I feel like the whole process has gotten out of my grasp and I have to figure out ways to lock the creative process down some so I’m still in control. All this back and forth between myself and AI causes me to feel like I’m spinning my wheels.

I’m not getting any younger. I still really want to wrap up this novel and start querying by late spring 2026. But…I don’t know. It could be fall 2026 before I get to that point.

Which is, I fear, kind of depressing.

I See AI As Like A New Form Of Word Processor When It Comes To Writing This Novel

by Shelt Garner
@sheltgarner

I am doing everything in my power to make sure that all the actual writing of this novel is done by me, me, me. But one thing is also clear — writing a novel is easy, writing a good novel is difficult.

As such, I am leaning into AI to help me with the backend of writing this novel. It speeds up and streamlines a lot of the tedious elements of development, like scene summaries.

So, in a sense, it’s just a tool when it comes to writing this novel. But, I must admit, it’s also something of a collaborator. I sometimes get really good advice from the LLMs that I use on a regular basis.

That’s why I finally broke down and paid for Claude LLM. It’s a great manuscript consultant and I like how it challenges me more than some of the other LLMs I use.

Anyway, wish me luck, I guess. It’s fortunate that I actually usually really enjoy — even love — the act of writing.

At A Loss (For The Moment)

by Shelt Garner
@sheltgarner

I have been using AI to game out the outline of my scifi dramedy to reasonable success. But, as always, I run smack into a problem when I take the wheel, as it were.

When I start to go through the outline and change things relative to my own vision of the novel, sometimes I really have problems figuring out what to do. That’s what’s going on right now with the first chapter which I, yet again, working on.

I really feel like I’m spinning my wheels, yet again, on this novel, but lulz, the novel is getting much, much better because my hero is becoming a lot more proactive. That’s been a real problem with this novel — and all my novels that I’ve attempted: my hero has been too passive.

I think that says more about how I view the world than anything else, but again, lulz.

Anyway, so right now, things are like this one chapter of setup and then the inciting incident happens in the second chapter. I’ve finished fleshing out the first three scenes, which is good. But the next two scenes are really causing me troubles.

As an aside, I’m really annoyed at how leery of sexual content all the major LLMs are as I work on this novel. One of them even accused me of being “lazy!” But, I think they probably had a point. I do tend slip into writing spicy scenes when I’m bored or can’t figure out what to do with a scene.

But it really is a pain to move the creative ship of state from what I’m given by the AI for an outline and turning into something that interests me, or I actually want to expend the time writing.

Wish me luck

Finished The First Draft Of The Scifi Dramedy I’ve Been Working On

After several false starts with other science fiction projects that never quite found their footing, I’m excited to announce that I’ve finally finished the first draft of my sci-fi dramedy. This one feels different—more focused, more intentional.

The concept emerged from wanting to explore the sweet spot between two films that have stuck with me: Her and Ex Machina. There’s something compelling about android narratives that I feel hasn’t been fully explored yet—specifically, the potential for a more intimate, relationship-driven story in the vein of Annie Hall. I’m not claiming to be anywhere near Woody Allen’s caliber as a storyteller, but that’s the general tone I’m aiming for: thoughtful, character-driven, with touches of humor alongside the deeper questions.

Now comes the traditional advice: set the manuscript aside for a month to gain perspective before diving into revisions. In an ideal world, I’d follow this wisdom to the letter. Unfortunately, my timeline is compressed. Life has a way of intervening, and I know that significant changes are coming in late spring 2026—right around when I hope to begin querying this novel. Given these circumstances, I’m planning for a shorter break: perhaps a few days, maybe a few weeks at most.

The practical reality is that I can’t afford to let this project sit idle for an extended period. Between the natural pressures of time and the knowledge that my circumstances will shift dramatically next year, momentum feels crucial.

For now, though, I have a stack of books waiting and a queue of films and shows I’ve been meaning to catch up on. This brief respite might stretch my break to a few days, or possibly longer if I get particularly absorbed in my reading and viewing list.

Either way, the first draft is done. That’s something worth celebrating.