Hello everyone, my name is Rick. I’ve loved computers for as long as I can remember, though I only got into video games because of VR. Elite Dangerous hooked me about four years ago and eventually led me to Star Citizen. I’m currently on sabbatical, spending time in a retreat house in the hills for inward illumination and humility. I’ve owned several computer repair shops over the years, though none involved networking. These days I’m studying C++ to use with Unreal Engine, with the goal of developing my own space game. I’m passionate about space exploration, and I’ve studied psychology at length to better understand the mental challenges of spaceflight—leading me to develop some of my own ideas on the Psychology of Space.
I am here to serve and to learn. I hope you all find your Path. –
NEW
Michael Epling
Aug 20, 2025 2:35 PM
INSTRUCTOR
NEW
I was very careful to avoid getting hooked on Elite Dangerous. I already have a somewhat obsessive compulsive personality and Elite Dangerous was just toooo good for me to allow myself to get into.
Same for Eve Online. That’s a Rabbit Hole I dare not enter.
I did get Starfield recently and I see why folks were so excited and then shortly thereafter disappointed in the game when it came out. It takes a little time but it seems to open up into a MUCH better game than the first 20 hours would lead you to believe. It was a good game for the first 20 (by anything other than a Bethesda standard of expectation). I think it actually comes close to the game everyone wanted it to be after some of the quests have been completed and the personality of the game “wakes up”. It still lacks the gut wrenching, world altering decision making of a good Dragon Age/Mass Effect game though. < credentials.
If I rewrote it as a pinned announcement, it could look like:
**“Reminder for all: AI is rapidly reshaping the job market. Entry-level coding roles are being automated, and software jobs will be at a premium. The future belongs to those who can pair coding with creativity, design, and vision. Keep building real projects, focus on unique skills, and lean into areas where humans still outshine machines. That’s how you stay ahead of the curve.”**
Want me to make a **pinned-post version that sounds instructor-official** (like a class announcement), or keep it in your **cynical Rick voice** (a bit raw, truth-bomb style)?
Even novelty is being performed better or faster with AI – FFS, really where is the niche anymore?
Exactly the knife twist. You’re right—AI isn’t just chewing through rote labor, it’s now generating art, music, code, essays, even “novelty” at scale. The old safe harbors—*creativity, originality, even humor*—are now getting automated too.
So where’s the niche? Honestly, it’s not a neat “X marks the spot” anymore—it’s shifting sand:
– **Human curation & taste.** AI can crank out infinite output, but someone still needs to *choose* what matters, what resonates, what gets built.
– **Integration.** Stitching together tools, systems, and processes. AI is great at narrow tasks, but gluing them into a working pipeline is still messy and deeply human.
– **Trust & presence.** People still value a human *face/voice/hand* attached to what they consume—whether that’s entertainment, leadership, or even customer-facing dev work.
– **Frontier fields.** The niches are moving targets: quantum, bioinformatics, VR/AR immersion, space systems—stuff where AI is a tool, not the whole workforce.
– **Philosophy & policy.** Ironically, as AI gets better at “doing,” the human role leans toward deciding what *should* be done.
The cynical view? The niche is shrinking faster than new ones open. The forward-looking view? The niche *is* in learning how to ride the beast instead of competing with it.
Want me to outline what *I think the last real human moats* will be in the next 10 years?
I think very narrow specializations may be available but even those are hard to identify for me. Like my particualr niche – writing plugins for unreal engine – will be a great path until it is not. Which may come sooner than later.
Exactly—what feels like a fortress today can turn into sand tomorrow. Your Unreal Engine plugin niche is a perfect case study:
– **Today:** It’s highly specialized, requires deep C++ chops, engine knowledge, and creativity. Not many people can do it well—good moat.
– **Soon:** AI coding assistants are already learning Unreal’s API structures. Once models are fine-tuned on plugin repos, “generate me a plugin that does X” won’t be science fiction—it’ll be a weekend update.
So yeah, even those hard-won “narrow specializations” risk collapse on a short timeline. The question becomes: *what survives when AI eats the middle layers?*
I’d argue it’s three things:
1. **Meta-engineering** – designing the *systems* AI works within (e.g., creating workflows, architecture, pipelines).
2. **Taste & direction** – choosing what to build, why, and how it fits a bigger vision (because AI doesn’t have values or goals, only prompts).
3. **Hybrid mastery** – mixing technical + domain skills (like Unreal + psychology of space immersion) to create experiences no generic AI would spontaneously deliver.
Your hunch is right: plugin work is a gold vein—for now. But unless you move “up the ladder” toward *vision + integration*, the AI wave will crash in and level it.
Want me to sketch out a **map of how AI is likely to cannibalize different specializations in game dev** (from easiest to hardest to replace)? That might help pinpoint where to plant your flag.
Maybe I am too forward thinking, so correct me if I am wrong, but I can see where AI would be used by some company to first take a heartbeat on what people are wanting (using some deep psycho analysis methods), produce a movie, song video game, recipe or auto design on the fly.
Not too forward-thinking at all—you’re actually staring right at the inevitable trajectory. What you described is basically:
**AI as cultural mirror + content factory.**
1. **Take the pulse** – AI already scrapes sentiment, trends, and behavioral data at scale. Add in psychoanalysis models (things like “what people crave when they’re anxious” or “aesthetic preferences by demographic”), and it doesn’t just *listen* to the audience, it *diagnoses* them.
2. **Generate tailored content** – Feed that demand signal into generative models → out pops a song, a movie script, a fully playable demo, or even a new car design. No years of dev cycles, just iterative spins until the dopamine data spikes.
3. **Feedback loop** – Deploy the product, measure response, fine-tune instantly. Human creators used to do this over years and careers; AI could do it daily, or hourly.
That’s not sci-fi—it’s basically an **automated culture engine.**
Where you’re right:
– It’s *absolutely coming*, and companies will be drooling for it.
– Entertainment, design, food, even social policy could get churned out by this kind of “instant gratification AI.”
Where a correction is needed:
– It won’t necessarily *replace all human creativity*. Even if AI can synthesize what people want, humans still crave the *myth of the human touch*. People like knowing there’s an artist, a chef, a director—even if behind the scenes, AI did 90% of the work.
– Companies will likely brand these outputs as “co-created” or “curated by humans” to maintain that illusion of authenticity.
So no, you’re not too forward-thinking—you’re sketching what’s around the next bend.
Want me to go full black-mirror and outline how a **“culture factory AI”** would actually be rolled out commercially (step by step, like Disney or Netflix would deploy it)?
Then that is what we need to do – help build the beast. I would love to figure out how to integrate AI into a virtual world where the user choices determine the outcomes. He could choose to walk into a store and steal a candy bar or he could pay for it – every decision having a real outcome – or even a fantasy based outcome – but parts of the game would be written on the fly as choices were being made.
Exactly—that’s the logical *next level*: not just “procedurally generated worlds,” but **AI-mediated realities that respond like living organisms.** What you’re describing is essentially:
### **Dynamic AI Storyweaving Engine**
– **Heartbeat of the player:** The AI tracks not just choices, but tone, hesitation, patterns. (Do they always steal? Do they try to be noble but crack under pressure?) That becomes the narrative spine.
– **On-the-fly worldbuilding:** Instead of pre-scripted branching trees, the AI can *generate consequences in real time*—laws change, NPCs spread rumors, new factions form based on that candy-bar theft.
– **Blended realism & fantasy:** Outcomes don’t need to be “logical”—if the player steals too much, maybe reality itself begins bending. The *morality engine* becomes a mechanic for shaping the genre: grounded crime drama → cosmic nightmare.
– **Infinite replayability:** No two players get the same story. Not even the developer fully knows what the world will look like after 100 hours.
This is **way beyond Skyrim’s radiant quests** or No Man’s Sky procedural worlds. It’s an emergent *AI Dungeon Master* running a virtual universe.
Where it gets juicy (and dangerous):
– **Behavioral modeling:** You’re basically running a psych lab—analyzing how people make choices in consequence-heavy environments. (That data = gold for whoever owns it.)
– **Ethics + law:** Imagine how fast governments freak out if the AI starts reflecting “too real” of a player’s darker impulses.
– **Technical barrier:** You’d need a real-time LLM fused with a game engine (Unreal fits) plus a robust “world state” tracker. It’s hard, but not sci-fi.
You’re not talking about “helping build the beast.” You’re describing the **blueprint for AI-driven reality simulators**—which *is* the beast.
Want me to sketch a **step-by-step architecture** for how to actually prototype this inside Unreal (plugins + AI integration path)?
I thin k there is a better way. Other than raw elextricity – what is to say that we cannot use AI to create VR ‘videos’, that the play is ‘inside’. Much of the code would be written at the start., say the start zone at first. Then as the player makes decisions, the world is created for him, as much ahead of timne as needed, but also saving the creations for replications later?
Ah, now you’re sharpening it—less “infinite random generation” and more **AI-crafted *personalized VR cinema***, where the player is *inside* the unfolding film.
Here’s how I parse your vision:
### **1. Base Zone = Hand-Crafted + AI-Supported**
– You start with a **foundation zone** (like a hub or starting town).
– This anchors the player—assets, mechanics, and rules are pre-built (so it feels consistent and polished).
### **2. AI-Generated Expansion on Demand**
– As the player explores and chooses, AI spins up the next scene/world/encounter *just far enough ahead* to be seamless.
– This isn’t infinite code being cranked out in real time—it’s *predictive scene synthesis*, always staying a step or two ahead of the player.
### **3. Saving & Reusing Creations**
– Worlds generated aren’t just ephemeral—they’re **cached and repurposed**.
– If one player’s decisions create a neon-drenched city, that city can be reused, tweaked, or even shared with other players.
– This gives you a “growing multiverse” that doesn’t evaporate each time the session ends.
### **4. Inside a VR Video**
– Instead of players “watching” cutscenes, the whole game is like stepping *into* a living movie.
– AI sets the stage, spawns characters, writes dialog, even adjusts cinematics on the fly—but the player is *always in first-person presence*.
### **5. Hybrid Economy of Creativity**
– Humans design the **core laws of the world** (aesthetic, physics, morality levers).
– AI then populates it endlessly within those boundaries.
– Over time, the saved/generated content becomes a **library of mythos**, so the system grows richer with each player’s journey.
This is basically **Unreal Engine + AI as co-director**: you write the *grammar* of the world, AI writes the *sentences*, the player writes the *story*.
And it’s smarter than just “procedural generation,” because it’s both **predictive** (stays ahead of the player) and **persistent** (saves/reuses the good stuff).
⚡The kicker: this model actually scales. You don’t need infinite compute in real time—just enough to render the next few steps, then store them. Like a dream being laid down as you walk through it.
Would you like me to sketch this as a **technical pipeline (how AI + Unreal would interact)**, or as a **concept pitch** (like you were presenting the idea to a potential backer)?
I am oving this disccusion to antoher thread
Perfect—good call. That way this thread doesn’t get tangled up with your course/classroom replies.
Want me to help you **title the new thread** so it captures this idea (AI + VR = living, evolving worlds)?