Yes, exactly. Inundate the world with superficially plausible yet hollow content, including any desired themes. People who aren't very discerning won't complain; the others will be outmatched and find that 99/100 pieces are all noise and they will need to spend increasing amounts of time trying to find the 1, if they can.
I think there are some good parallels with Amazon: the broken sorting and manipulated unit pricing, coupled with the avalanche of cheap clones pushes users to give up and just buy one of the top listed products (a featured listing/Amazon-clone). If you do a web search for various products and go to images, Amazon product links often take up 50-90% of the results.
But the dopamine descent require strong discipline to stop there, and most don't .
The other is creating multi-modal models with a better understanding of our world. LLMs often fail at incredibly basic spatial reasoning ("someone left a package in front of your apartment, describe going there", or the "should I drive to the car wash or go there", etc). World models excel at these kinds of things (in theory). They develop a great understanding of physical spaces, object interactions, etc. They can simulate fluids, rigid body physics etc. You "just" have to get really good at making world models, then somehow marry them with an LLM in a way that ensures the LLM can benefit from the world model's training data. Nobody has managed to really do that yet
So lots of hopes for the future. Until then they get commercialized as video models, or ways to experience your favorite forest, or to have a really bad video game ... whatever can be sold on a short time horizon to finance the actual goals
There are a lot of areas where predictive models make sense in the robotics stack, but doing it with "video world models" as is trendy this year is likely a bet in the wrong direction according to the evidence we have been amassing in the last 6 months.
Making a world internally consistent by explicit placement gets harder as you increase in scale. When internal consistency is a factor impacting quality, there is a scale at which generated content eventually becomes the higher quality solution.
Secondly, when generating content with AI, the same rules around carelessness apply. There are certainly generative AI tools out there that offer few options when it comes to composing what you want, that is not a necessary part of AI, some of it is because people are wanting rudimentary interfaces, some of it is that the generators are sufficiently new that the control mechanisms are limited because they are focused upon doing something at all before doing it highly controlled, in some ways the problem is that things are new enough that it can be hard to describe what is desirable controllability, making the generator to see what people would like it to be able to do is, I think, a reasonable path to follow prior to creating the control that people want. Part of it is also that there _are_ tools that give a high level of control over what is generated but far fewer people get to see them. There are ways to control styles, object placement, camera motions, scene compositions, etc. The more specialised you get, the smaller the subset of people who need that specific control.
I think AI can make things possible for people who could not have done so without them, but it's still going to take care to make something special.
It seems to, even.
Whereas if you hand a router to someone with a flush trim but in it and ask them to clean up the edge of a table they will take one look at it and nope away from that dangerous spinning thing.
If they have the mind to give it a shot and despite a quality tool and bit they bite into the table and ruin the line (or something much worse) no one will be surprised—-they have no experience or recognition of what expertise is in woodworking.
But with AI, it is much more hazy what expertise is.
The methodology for quality results is changing each week and the articulation in personal tooling involved makes it challenging to adopt another “expert”’s workflows.
Assuming that the LLM search will be meaningfully better at cutting through bullshit than the generating model was at avoiding creating it is, charitably, dubious. Assuming that it won’t be every bit as gamable as Google results is as or more dubious.
One aspect of intentionality is that there’ll be a narrative payoff when you find something you find interesting. In videogames, the world is mostly pre-designed, so the designer has to predict what you’ll be interested in for the most part (In pen and paper RPGs, this is usually done better, because the human dungeon master/DM can plan ahead, but also improvise a payoff or modify the plot between sessions). If there was a world model generated game world, I guess the model would have to be “smart” though to setup and execute those payoffs.
An advantage that the world model would have (and shares with a good human DM) is that everything is an interactable, and the players get to pick what they think is interesting. If everything is improv with a loose skeleton around it, you don’t have to predict as far out. I think world model generated games, if they even become a thing, will be quite a bit worse than conventionally designed ones for a long time (improv can be quite shallow!) but have a lot of potential if they work out.
FromSoft is an interesting example. They make the game more believable by having extremely missable quests, just, most of them don’t block progress through the game, and you usually stumble across enough side quests naturally (although IMO the density was too low in Elden Ring, their system showed a bit of weakness in the less-guided context). The plot is pretty vague, but the vibes tell enough of a story that you don’t really mind. It’s sort of improv/pen-and-paper but the player’s imagination is doing the job of the DM.
Fromsoft is perfectly happy for you to miss all of the direct exposition. It's as they intended and most people do. The intentionality of their world still draws people in and gives the world a sense of groundedness that keeps people coming back and separates it from the pale imitations. It's more than them being good at 'Vibes'.
The environment is built on the bones of a greater ongoing narrative that is intentionally obscured, even from the player who reads everything.
Dark Souls is a world in a constant cycle of Rebirth, Decay, the struggle against Entropy. All civilizations, at the end of the series, are stacked one upon another in an endless expanse of ash and dust as you bear witness to an permanent eclipse, a fading star, as time itself dissolves and the last fire fades.
Before that heavy handed stuff though, the simple matter of the direction your character travels reinforces these motifs. Down to the deepest depths and you'll witness what remains of the first civilizations. Climb up and you see the desperate attempts by the powerful to impose a false order that they hoped could forestall the inevitable.
You can even shatter the illusion of a golden order in the first game if you find the extremely missable secret boss. It couldn't be any more clearly 'said' if you were interested in paying attention.
Adding an AI model to explain or 'improv' the story of the world would destroy the whole purpose.
That's a pretty specific and one-sided example. There are tons of good games that don't rely on elaborate item placement (e.g. many Bethesda games are great because most items are useless decorations, they broke that rule in recent games, giving the purpose to clutter, and it made them a lot worse). There are tons of good games not relying on this intentionality at all, they're either literally random cool ideas thrown at the wall, or even procedurally generated.
Where you look for an intentionally evoked experience authored by a game designer, I am looking for an unexplored world unfolding before me filled with emergent and unique phenomena that perhaps no one and not even the game designer has seen before.
These world models are key for robotic and coherence in video generation.
Give a world model images of a factory, the robot now can simulate tasks and do the best result.
Give a world model images/context etc. and it can generate a coherent world for video generation.
What this world model system might be able to do for us in regards of gaming or virtual reality: Either simulate 'old' environments like the house of your grandparents (gaussian splatting but interactive) or potential new ones like a house, kitchen, remodeling.
It can also be a very interesting easy to approach VR environment were you can start building your world with voice. That would be very intentional. After all world building is not necessarily connected to being able to generate 3d assets. Just because you need to go this route today, doesn't mean you have to do this tomorrow.
for example, I am 100% certain that ANY model could write a better Dragon Age sequel than the rotting corpse of Bioware did, because only humans can despise their audience and their source material. an LLM would dutifully attempt to produce more of the thing rather than 're-imagine' the thing for 'the modern audience'.
Many of the most popular games in the past decade are procedurally generated and have nothing “intentionally” placed (apart from tuning/tweaking the balance of the seeding algorithms).
I think you underestimate the intentionality that goes into developing procedural generation. Something like Dwarf Fortress isn't "place objects randomly" - it is layers upon layers of carefully crafted systems that build upon each other to produce specific patterns of outcome
I guess what I'm saying is: Couldn't a world model with targeted training and thoughtfully tuned system prompts be directionally similar to the layered systems to produce specific patterns of outcome?
Are video game developers using these systems in their workflows? Would love to learn more!
The combination of "many", "most popular", and "nothing" is overstating it by a wide margin but for example the majority of the vegetation in games as far back as oblivion was procedurally placed.
Yes, we haven't gone that far with creating consciousness yet, but there is gonna be a lot of money around neural computing devices for consumers in the coming decades, so that will speed up knowing what sense data you need for consciousness.
Wait until you learn about what we do to chickens.
You’re right - but that world is not the end of the story. The intentionality matters. Human creations matter because they connect us. I don’t know how long it will take, but people will build judgement as to what makes for good use of these tools to make meaningful things and expand our creative horizons in deeply human ways. Mind you, there will always be shallow slop. It’ll just take time for creators to learn how to use these tools to make something that isn’t slop.
On that first point I think it's important to remember that the lineage of video games comes from board & card games and sports. There's always been an ability to inject more complexity and less-intentionality into those things. Sports in some ways are like a simplified and altered role-play of war battles, and more realistic war roleplaying does exist but it has less appeal.
As humans we like solving things and noticing patterns and the intentionality of games taps into that appeal.
On the latter point I do think these world models will eventually be used to meaningfully contribute to building games. I think people will have to find new ways to design that balances intentionality against the freeform nature of these simulations, but it may take a while to have the capability to do so.
Like for instance... Dwarf Fortress? Minecraft?
Generative AI is just another method to go procedural generation. Not necessarily a better way. Or you could even argue that procedural generation is a form of generative AI... But either way, there are games where the lack of intentionality is central to the appeal.
its a very dead game on its own. they are still very intentional about adding and changing the tools by which you make your own fun though.
DF for some doesn't fit into this category for me. Minecraft feels dead to me, while many other games that utilize procedural generation are not.
Put another way, the average game quality will go down, but the actual rate of "Great" games will go up.
I take raw material and make something out of it with a circular saw, largely unrestrained by anything other than cost, skill, and material.
With a microwave, I make things hot so I can eat them.
Aside: Also, I wonder why that is? Why do we regard the microwave as "degenerate" compared to the oven? Why is baking seen as a calling while microwaving is, well, not? Is it the ease of the microwave makes the effort less impressive? Maybe it is that you can't achieve certain effects like browning? Is it because of it's 1970's association with "radiation" and tv dinners? Is it just cultural inertia?
Proper microwaving is what gave rise to the entire concept of "fast casual" restaurants, famously AppleBees (or "club B's" in the late night focus iterations!)
Complex entrees that could be partially cooked and frozen. Then rapidly microwaved on a custom program that varies the timing and intensity of cooking. Then finished on a grill or conventional heat source for less than 1 minutes.
Microwaving food generally produces a lower quality finished product. But you can take a similar approach at home. The short cut is to just double the cooking time and cook at 50% power. Then throw whatever the item is in a preheated pan for about 1 minute if it's applicable. Other variations are possible too, I air fry finish most things like chicken nuggets, tater tots etc and the difference is considerable while still offer a significantly reduced cooking time.
I have used none of those recipes. The microwave is for making cold pizza 10% more palatable (or 80% more palatable if I've been drinking). In that regard, the LLMs are microwaves really works for me: if I'm using one I either I want something fast and casual, or I'm drunk.
Of course you do! Please take the time to upload and share that's an amazingly niche knowledge pocket.
I think it is interesting (though I only partially agree) that microwave meals require standardization to scale. Let's say that was true, why couldn't a modern microwave have a small camera and a set of heuristics for how to cook just about anything by turning the gun on and off at particular points when it recognizes a food? Maybe without intelligence, a microwave does need standardization; but we can put intelligence (ideally offline) in just about anything these days?
I wonder if with sufficient control if a microwave could ever brown? I wonder if it could reliably bake?
https://www.youtube.com/watch?v=2tg3-93jKvc - Chicken Good!
So partly due to that, I think if you wanted to do full robotic cookery with a microwave beyond the TV-dinner stage you'd still need control over the other objects rather than just switching the magnetron. Stirring for even heat distribution is another thing.
And then it's separately interesting how, in a really skew way, “deliver heat using RF energy” came back from a different angle with the rise of induction stoves!
Why do I need more slopware? I have an entire Steam library of excellent games that deserve to be played first.
In reality, I don’t see any of this trending towards the theoretical happy path everybody always talks about. Most people give up trying to find something good on Amazon and just buy whatever vaguely plausible knock-off garbage shows up in the first few search results. Most people just take any job interview they’re offered even if it sucks. Most HR people don’t use it to enhance the quality of their decisions — it replaces their decision-making roles in many respects.
I’m an art school graduate and talk am in many art discussion communities. This is causing a massive industry-wide morale crater. In any sort of art, it damn near eliminates the reward of craftsmanship in favor of marketing useless trend-of-the-week bullshit. Far fewer people enter a market that can’t sustain them. The idea that this is going to create ‘more artists’ and therefore that must mean there must be more skilled artists is fantasy. The skills you learn by prompting are not even on the same track to learning how to create things yourself. You essentially become a high-school intern acting as an art director, commissioning pieces. It’s instant gratification for people who don’t care enough about something to learn how to do it for real.
I don't think these will create "artist" in any sense, but I do think it will lower the barrier dramatically for people creating games. Most people will interact with it like Lieutenant Barclay interacting with the holodeck, doing little more than wish fulfillment. But I think a few people will be able to interact with it in ways that create art.
In no way am I implying that the net net of AI will be good for humanity as a whole (I think that is too big a question), but I do think the power of World Models will probably result in a far more people being able to say "I have created a game".
I honestly don't have anything useful to say about what LLMs are doing to many human fields. I can understand how frustrating it must feel to see LLMs demonstrate superhuman "skill" (I don't really think they are skilled) at orders of magnitude less cost than a good artist. It isn't just that they don't seem to innovate (only permute), it is that they will literally take even the tiniest bit of creativity and novelty and immediately fine tune and create derivative works on any idea at scale. I can see how that might really demotivate any desire to push the boundaries of art for any human being.
Artists make thousands of tiny decisions when creating art— many of them unconscious. It’s true across mediums— e.g. digital art, oil painting, collage, and architecture photography— and even realms, like music, visual art, and animation. Learning to make art is a process of learning to make those conscious and unconscious decisions to create something that communicates what you want it to.
AI simultaneously makes it harder to make those decisions, and supplants those decisions with the amalgamated decisions of other artists. The skills to do it are fundamentally different. It’s not an incremental change like physical media to digital media, which requires the same exact underlying set of cognitive skills and processes even if the logistics were different. Novices love AI because they don’t have to learn how to make those artistic judgements, or the craft of implementing them… unfortunately, learning that stuff is learning art. Using AI to generate very specific images can be a difficult, deep, and creative process, but it is fundamentally different than creating art— it is art direction. You could definitely consider a meaningfully curated collection of images other people (or machines) you generated ‘art,’ but no matter how closely an art director micromanaged an artist making a piece, it’s still, fundamentally, the artist’s low-level sensibility, experience, and perspective on the micro level that made that piece what it was. If that art director put their name on that artist’s piece, they’d justifiably be run out of town with pitchforks.
But AI does not make learning art more accessible to people. They make it nearly free to commission art from a computer. This isn’t a purely philosophical distinction — it would be like someone saying they’re an experienced driver because they use Waymo cars. Even if they get a new ride every block to make sure they take the exact route they want to take, that doesn’t mean they’re driving. It’s just a different thing. And if you’re banking on more Waymo cars making it easier for people to get in cars, and increasing the number of car trips to increase the number of experienced drivers in the world, that’s a fundamentally flawed way of reasoning about it.
Would you consider it possible that the way non-intentionally placed items break the game immersion for you is because they appear in such a way that you think you can interact with them in a certain way, but you can't?
Like if there's an extra door in the house you're trying to get into, but that door doesn't really open, then in your mind that breaks the integrity of the game's systems. If so, I think the LLM response is that there are no more doors that don't open and that the world can be generated as needed.
No computer can handle the complexity of even a small town. But it would be possible, at least in the future, to generate the part of the world you interact with, which would heighten the emersion.
the intentionally placed tree serves no particular in-game job mechanically. it instead points your eyes to the right place when you walk up the path, and then again when you look back down from above.
when they're saying everything is intentionally placed, they mean everything, whether it looks important or not. It's all directed to a cohesive core
It sounds like you might be assuming intent where there is little. I suspect if you didn't know a poem was written by an LLM you might ascribe purpose and intent to many aspects - rather like a high-school English language teacher.
LLMs had nothing to do with this
It’s been my belief for several years that this is how the future of games will be constructed. Data in the background, game engine for rules application/ physics execution/orchestration/maybe low-poly rendering, an AI world model taking low resolution input in generating customized visuals/effects/textures/everything, even camera location, but still constrained by concrete rules in the game engine.
I’m certain one day it might all be handled by AI, but the above seems much more realistic and achievable that expecting AI to do all of these things, at one, correctly, every frame.
So, what's the deal?
As with ANY work in life, the quality of the result is a direct reflection of care and intention behind it. Simplified, it's a reflection of how much _you_ put effort in it. It always shows. Even in AI day and age. It's just that path to a result (without effort) is now way shorter so volume is showing up and diluting the overall impression. The latter kind of cheapens every field it touches, so even more effort will have to be put in to show up on the radar.
They can be very powerful tools but creating meaningful art/subjective work that is actually good is actually a borderline impossible task for LLMs since it requires genuine creativity.
Similarly, I am seeing coding agents be infinitely more useful to people who already code than non-technical people thinking they can vibe code their own Salesforce.
Its just that AI doesn't work well as a "productivity booster" tool in the marketing context. So they are pushing towards the idea that anyone can do anything with AI which is imo a really stupid hill to die on or base your ipo on.
Huge percentage of AI spend is on this with the gamble people won’t care if it’s less humane they’ll consume it anyway.
I'm not a game developer myself, but some of my favorite games carry a deep sense of intentionality. For instance, there is typically not a single item misplaced in a FromSoftware game (or, for instance, Lies of P -- more recently). Almost every object is placed intentionally.
Games which lack this intentionality often feel dead in contrast. You run into experiences which break immersion, or pull you out of the experience that the developer is trying to convey to you.
It's difficult for me to imagine world models getting to a place where this sort of intentionality is captured. The best frontier LLMs fail to do this in writing (all the time), and even in code, and the surface of experiences for those mediums often feel "smaller" than the user interaction profile of a video game.
It's not clear how these world models could be used modularly by humans hoping to develop intentional experiences? I don't know much about their usage (LLMs are somewhat modular: they can produce text, humans can work on it, other LLMs can work on it). Is the same true for the video output here?
All this to say, I'm impressed with these world models, but similar to LLMs with writing, it's not really clear what it is that we are building towards? We are able to create less satisfying, less humane experiences faster? Perhaps the most immediate benefit is the ability for robotic systems to simulate actions (by conjuring a world, and imagining the implications).
In general, I have the feeling that we are hurtling towards a world with less intentionality behind all the things we experience. Everything becomes impersonal, more noisy, etc.