Wednesday, September 23, 2026

Fall 2026


This year has been another roller coaster ride. Starting with Agents. Looking back at the plan I created for STORiCORE around this time in 2024, agents were something I anticipated might be in the not-so-distant future. It took over a year, as most of 2025 was gradual AI improvements until December, when Agents began to improve enough that I could actually use them and reconsider the path ahead based on what I had accomplished without agents in 2025 and what I could do going forward with agents.  

I am not an AI expert with 20 years of education on all things computer programming; I am a storyteller. And AI is something I have seen as an evolving toolkit. My plans for using AI have evolved since 2023, just like AI has evolved.  

I am an artist who puts time and care into the stories that I write. I place great value in those stories, regardless of whether others do or not. Why? Because I write the stories I want to read and see on a screen. 

Stories that you and I will soon be able to interact with. Again, what makes the stories I write something that are worth interacting with? I have learned not to presume that anything will be a big success. But I will be leading with my favorite stories that I have written over the past 30 years. 

A long time ago, I started writing spec screenplays. Stories of my own creation that nobody paid me to write and stories that weren't meant to be fan fiction, not something overly similar to what was successful before, nor were they episodes in someone else's TV shows. 

I have always written what I want to write, and while I write stories for other people as well, they are not the main reason I am writing the story. It is about taste. I write what I am inspired to write and to my taste and not to that of others. 

Example: You don't write a superhero story that is exactly like another story and just change the names because everyone can see the rip-off. Plus, superheroes? Barf. I remember going to REDBOX back in the day and seeing the knockoff versions of the A-list movies of the same storyline already in the box when the A-list version was still in theaters. That is the worst kind of filmmaking in my view. Making a shittier version faster to steal a few bucks off of mistaken identity. No refunds. Ha-ha, sucker. Trash.

I like stories that challenge the reader or viewer in some way. Not that I like them to be so edgy and raw that only a handful of people would ever want to read or watch them. No way. I have to want to read or watch them myself, and I want to be entertained and challenged in some way. Whether that is emotionally, intellectually, psychologically, spiritually, etc. I want to feel something, and I want others to feel something as well. Fan fiction is fine; social media is loaded with it, and that is fine for those who like it, but I want to see something new or at least inventive and clever. Or at least something that doesn't feel like a shittier version of something else. Similar but different, yes. Not the same but worse.  

Okay, you get it. Enough about why I write. 

In August of 2024, I thought apps were the way to go to raise money. I did not think I could raise money based on how I could use AI to help me creatively, largely because video was ok but not believable to me. I could not focus on what I had laid out in the TV series I had just failed to garner interest in over the past few months; my plans had that tech taking at least until 2025, when it actually took until late 2026. I had the vision for the future as clear as crystal in my head. And as of today, all that I laid out in the show bible can be achieved by myself with the help of AI. It could not in 2024 or 2025. 

Back then I wanted the story to be filmed like a traditional TV series. Now I couldn't care less. Not because I don't prefer traditional filmmaking, I do. But no one is begging for my stories to turn them into the movies and shows I want to see made. Am I supposed to continue to think the way I used to? Do I think it is smart for me to continue writing spec scripts and then contacting people to see if maybe, just maybe they might actually read my email, take a real interest in the story, ask for the script and the bible and the treatment, then give me a call or send me an email that says they are interested, wait for things to come together, get paid, give up the story for others to do a complete rewrite, get delayed, finally get made, be furious with their choices, watch the public's reaction, be deemed a failure or a success, and then go again? I would love that. Call me. But the odds are even less in my favor now because more people are turning to AI. A real Catch 22. 

That was the plan for decades. It was humiliating and thankless, but I kept grinding because the stories meant something to me. And when something means something to you, you will move heaven and earth for it. Granted, it was more of a hobby for me because I never landed that big deal that would set me up where I could pitch any story idea and have it get optioned and made with a little less muss or fuss. I had accepted that, and to do so I had to lower my expectations and take satisfaction from creating the story and having lived in that world without it ever seeing a screen. These stories are a reflection of the times they are written in, and they are how I make sense of the world as I have passed through it. Could I have raised money to use AI to work on my stories instead of apps? Maybe. Possibly. But if I had, I would not have learned what I have since August of 2024. 

It was a bit foolish to think I could create three great apps right out of the gate that people would pay for, so I could then be able to tell dozens of my stories in different formats once the tools improved enough. That was the plan. The plan fell apart because I had never built an app before and the tools were slow to progress in 2025. Not to mention the fact that apps are not really a thing for builders because they all build their own tools now that the agent boom has been going on since last December.  

My view of agents has changed since the beginning of the year. I thought there might be two versions of agents I would work on. One where there is an agent market where a consumer might pay to have access to an agent for a certain task or tasks. Kind of like creating an app or hiring a worker. But in reality, the models and their harnesses have blown past that. People will continue to buy agents as tools and Apps. But it is the second thing that is dominating: the individual user being able to create their own app or agent workflow. This is what I have been focused on. I really thought I might create an agent for people, but what they really need is a personalized workflow just for them. The Twitter AI crowd is building for their own situation. TBF, they have been doing this for years, but it is the fact that everyone is showing off all the amazing things that they can do. It is its own podcast community of like-and-subscribe zombies whom I happen to like to watch. And I think this is the apex of working with AI so far. Creating pipelines and workflows that work for you. You can either buy the app or build the framework that works for your situation. The momentum is going in one direction, but the need for applications still remains. 

The benefit of creating your own software is that you own it. You run it internally. You can share access to it. But it is yours to change as you need. The bar has been lowered for people like me to be able to create apps and my own internal pipelines and workflows. And if I can do it after a few years as the technology gradually improved, with all the big labs talking about RSI (Recursive Self Improvement) having been achieved, the liftoff will be even more rapid now. That, regulatory capture, the first security issues to really start raising alarms globally, and a political push to get on the side of fewer data centers and safety have all created a moment of consternation for "the public". 

Regardless of what happens with AI, we aren't going backwards. The gains that have been made will remain available to the public because even if big companies pull them from the public, open source companies will have had time to distill their own open source models that can do most of what the other models did at 1/2 the cost. That means everyone can use AI to create whatever they need that an AI model can create at very low cost, and the trial and error period that I learned about in 2025 will never exist again. Why? Because the models are so good now that the trial and error that I dealt with in 2025 is gone. Things are easier. Therefore, people will not need to learn all the lessons I had to learn over a year ago. Smarter is better, but there are still mistakes. Just less than before, allowing actual use much faster and less need to dive as deep into understanding AI because AI understands us more. It understands everything more, and while it is not perfect, the need for you to be perfect has dramatically lowered. Can you feel the acceleration? Because it is here. 


I abandoned the big push to create all three apps because of agents. Yes, agents made it so I could more easily create the other apps, but where was the interest? There was none. And I am one person trying to do everything that I had never done before. It made a lot of sense to focus on agents and creativity as opposed to apps and a software company because I have enough content that once I get cranking it out, I can keep it flowing for years. And that means more to me than constantly improving multiple apps. 

That's the hope, at least. As I wrote last month, things are coming along just fine with the creative pipeline I created. 

Two things in the pipeline that I have put off were video and video games. Yes, Seedance 2.0 was amazing and came out in February, but I have been reluctant to use Chinese models for anything, but I am less concerned with that as far as video. However, most US models are not nearly as good as the Chinese models, and Seedance 2.5 just released, and it is even better than 2.0. 😶

Video is and was the one I cared about most because I have written dozens of screenplays for stories intended for the big screen &/or for TV screens. Video has gotten to a point where it is almost indistinguishable from real video. Animation is solved. Seedance has crushed it, and there are others and new world models like the Runway's Solaris that are making AI video far more appealing to me. The push for world models is massive and seems close to being solved, if it hasn't already. Cost is a big issue I have worried about with Seedance and others, but the prices have come down a good bit. 

In some ways, I wish I was starting my research now instead of winding down 3 1/2 years of research. I have gained so much valuable insight during that time and achieved more than I am probably willing to admit. The tools are so much better than 3 1/2 years ago. 

Would it be better to have waited until now to jump on the AI train? Maybe because I would not have had to do as much research as I have done because the tools are so much easier to work with that I maybe could have avoided the deep dive into ML, Deep Learning, Python, and the other courses I took in 2023 to try and understand LLMs and Generative AI. But then I wouldn't be at the point I am in my research and building. I am someone who likes to be able to take things apart to try and fix them when they break. I cannot physically do that with AI, but I have gained enough knowledge and confidence to be able to fix some of the things that AI, or myself, may break along the way.  

Time and money are once again my biggest impediments to achieving what I want to achieve. Yes, I can do much of the work in my pipeline for low cost, but creating movies, TV shows, and games will still cost me a small fortune and take up a lot of time. That said I am seeing things with Seedance 2.5 that make me believe that the abilities are at or very near where I wanted them to be back in 2024 before I was willing to jump back into video. 

Then OpenAI released Astra, and the number of video examples I have seen of people using Astra with Blender and Unreal Engine to create killer games, and it got me dreaming again. But it wasn't a dream at all. It was a realization. I realized that I could now use AI to complete every creative branch in my creative pipeline, with the ability to create 3-D worlds and games. The ability to have interactive games is massive. SORA was the one that got me properly dreaming because I could see the future through its imperfect videos. In creating the bible for the TV series I was pimping out back in 2024, I thought we could do all of it that had not actually been filmed for the TV series by summer of 2025. I was again a year off. But we are 100% there now. 

I could do the entire series without filming the TV series and doing the whole thing with AI. Which was what I wanted to do starting back in August 2024. The cost is the problem. 

It was even more of an issue before I created a pipeline designed to save money while allowing me to create one story in several different formats. Granted, Video and Games are still the most expensive, but once I implement them into the pipeline, I think I can cut the costs for them as well. Astra has proven it can help create Games and Videos. So can Fable, Opus 5.5, and GPT6 Sol. The last two within the past day. And that list will continue to grow as time goes on. 

That type of help in the process will be huge and save a lot of time. The cost again is the thing. It is becoming harder and harder for me to keep pushing Games and Videos with such great advances, but it still makes sense, considering I can do so much other story work just with a $20 subscription, and creating an AI movie would cost thousands. I suppose I could create an initial version of the game first to save money. It would cut down on the takes and just let a user run through the created world based on my stories. 

Quality was the issue with the old way of doing things. Not as much today. Cost has always been an issue with video. It is still an issue today, but there are ways to save money when working in some of the mediums. But there are still pricey subscriptions that users must have. The good thing is you do not have to spend MAX money every month. Only when you are working in a medium where you need those more expensive subscriptions. 

I think that building apps for others based on my ideas and then having to update them in my free time is not the way for me at this moment. You can whip up an app quickly these days. You can whip up an agent architecture that can handle a lot of work. To do it all as a solopreneur is when the expense starts to kick in. You can 100% create and run a business on your own, but it will cost you not just money but time. Agents help with time, but they cost a fortune for video or games or a lot of the work you may want to use agents for. So, I can't venture back down that path right now. I am jealous of these content creators who spend tens of thousands a month to test the tools and create content, but that is not me... at all. I have a set of creative objectives that involve storytelling and then a desire to use AI tools to help me deliver them to people. 

I thought helping people by creating apps that they might find helpful was a good idea. And it was, but not for me as a solopreneur in the year 2025. If I had waited a year, then I might have had a different experience and opinion. 

Even if I were to raise more money, I would not focus on creating apps for others but instead pursue my own creative projects. Focusing on creating apps to help others is a great feeling, and I had quite the adventure last year. Hell, I am sitting on two apps that I realize it would just be too much of a time suck to release and maintain on my own. If I had an endless supply of compute at my fingertips, I would spin up all sorts of agents, and maintaining things would be a lot easier as a solopreneur. But without a huge bankroll, it makes 0 sense. Now they are side projects that I may or may not release one day. More than likely, I will just shift my focus to something else.

I have to try things and see how they work to truly get an understanding of them. With AI, you have to be on your bike to keep up with all the updates and changes. That's why I have focused on the creative pipeline. I am not creating something to help others. Instead, I have created something to help me entertain others. 

I am working on everything in either Codex or Claude Code, because I can connect all the tools I need for app creation, work on my creative pipeline, or anything else I am working on. My budget is a lot tighter than it was this time last year, but that has meant I have had to be clever about how I use the tools. 

Speaking of money, over the past week I saw these two AI video influencers talk about the cost of creating their AI videos. One spent a small fortune, and the other mocked him and said he had done his for a fraction of the price. Either way, it is going to cost 10s of thousands of dollars at least to create a movie, with the high end coming in closer to 1/2 a million bucks. That seems too high, but those are the numbers each provided. A TV show would cost even more. Now you can see why I have been reluctant to dive into AI video. These videos are cool and all, but I need that cost to come way down to implement it into my pipeline for my movies and TV shows. Unless someone wants to help fund my first feature. I'll give you an executive producer credit, sing your praises, and split the profits. 


That is why, during the writing of this blog, it dawned on me that creating one of the games I have had in mind for the TV show I queried directors and producers with back in 2024 is starting to make more sense. It is the most ideal 1st-person video game that I came up with for the series. It is more hardcore than the other interactive experiences I had planned as extra transmedia experiences for the series. That or the fictional podcast may be first on the list. We'll see.

It all comes down to time and money. The pipeline for comics and cards is all about time because I don't need to spend thousands of dollars like with video. That video game and the podcast for the TV series fall somewhere in between comics and video. I am not an influencer or a well-funded creator, so I am ok not riding the wave of new capabilities if they are overly expensive. You have to do what you can and dream about the rest, with hope that one day the cost decreases enough or I can raise enough money to make those movies and TV shows. I can run video tests in the meantime just to stay up to date. 

No comments:
Labels: , , , , , , , , , , , , , , , ,
Subscribe to: Posts (Atom)