> movies are going to become video games and video games are going to become something unimaginably better
God what an annoying thing to say. Suggesting that movies will essentially become games, or that future video games will be unimaginably better (presumably because they are driven by generative world models) betrays such a surface level appreciation for what makes art and culture and all this humanity so beautiful as it is.
The entire point of movies is the deliberate choices of the filmmakers! Movies will not be improved by our ability to choose what the main character looks like or add a style filter. Super Mario 64 is great because it is an expression of the very hardware that constrains it. Inevitable future versions where anything can happen because someone hooked the ROM up to a generative world model won't be better for it.
I agree that this technology is absolutely world changing, and am cautiously eager to see what comes of it, but it just bums me out, perhaps needlessly, that the people at the helm can be so artless, unreflective. Whatever.
Sam had OpenAI start the Ghibli style image-gen (movies that are specifically loved for their artisanal quality). The man has patternly failed to read the room around culture.
And they will be so severely targeted to you that you won't be able to identify that they are designed to influence your purchasing or voting decisions. And they will be of production quality rivaling the highest budget films - rendered on your own hardware.
The gameplay _becomes_ the ad. If you have a full sash of potions in every slot, they spell "C","O","K","E". The monster-bashing sounds sound like onomatopoetic "Apple!" "Goog!". The armor colors are the same as the NFL teams nearest your locale!
I submit that most people haven't been able to do that for some time now. Advertising wouldn't be such big business if it didn't work on a macro scale.
I mean, sure, some advertisers might get more sophisticated. The garbagefest of internet advertising won't go anywhere, though. It'll still be as bleak and obnoxious as always, because someone will always accept a buck to let you barrage their visitors' eyes and ears.
Sounds like the plot from the movie matrix. Only difference between the movie and our real timeline - the ai didn't need to enslave humans, we willingly chose enslavement by ai for ourselves by continuing our google subscription.
It also sounds like Ready Player One, where the real world is so shit, everyone spends all their time in VR, using rented equipment, making nearly everyone a slave to the corporations.
Ultra nerdy point: in the matrix it’s revealed that humans did in fact choose to join it. And the meta point was that most of them still would choose it over reality [1] every day.
[1] reality is also somewhat in question in the films universe
It's from Matrix Reloaded, when the architect explains past failures of the matrix and that they only succeeded when they figured out that they had to offer a choice, even if that choice was subconscious. That nearly everyone accepted the matrix, and the few that did not escaped to Zion. That's the whole point of Neo, to manage the outliers and keep the system going.
At least that's my recollection. I've watched the matrix a few times :)
There was a guy named Cypher who betrayed the humans and cut a deal with the robots to live in the matrix as a rich guy. He fully accepts the steak he eats is not “real” but he enjoys it nonetheless
People already do this in many video games. The simulation isn't perfect, and maybe there's a tipping point where the simulation becomes real enough that it (maybe) supplants "reality". In a sense, this is disintermediation: instead of actually riding horse, I can ride a virtual dragon. Does that make it worse, or problematic? Again, we already have this, for driving race cars, fighting in wars, etc. etc. It's just a matter of fidelity and immersion?
It will lead nowhere. Graphics intensive video games as interactive environments benefit from efficient rendering done by hardware a fraction of the cost of what a neural network requires for the same output. This is a business dead end.
Instagram Stories were radically different in 2016 than the reels of today. They were far more personalized (less algorithmic public content), slideshow-like updates. Ten second max video clips, with most being just images. Ephemeral as well, where they'd get removed in 24 hours.
Many kids were, sometimes. Also, Vine was still pretty different from modern short form content, being capped at only a few seconds of length. Now most people consume tons of short form content, with many being 30s or more. It's pretty radically different in terms of what most people consume.
I for one would love to be able to experience Godzilla Minus Zero interactively. I get that it's different from movies. I remember seeing Avatar in 3-D and realizing that James Cameron was directing the audience's attention to a particular part of a scene by the way the 3-D focused, and it was a little off-putting at times. That's a small thing compared to what will be possible in a few years, but still -- being immersed in Avatar will be an amazing experience. I'd also love to see a version of Avatar written by '80s James Cameron -- he was a much better writer back then -- or maybe other, more skillful writers had greater influence on him? In any case, it's inarguable that Terminator, Aliens, The Abyss, and Terminator 2 are all better written than the Avatar films.
More and more off-topic, but this made me wonder: this interactivization would work much better for some works than others. Many comedies are scripted so tightly that I can't imagine any wider world around them (also, on the other side, Memento), while something like Twin Peaks would be perfectly expandable into a world (if you somehow had David Lynch in a datacenter, which sounds oxymoronic).
> I for one would love to be able to experience Godzilla Minus Zero interactively.
So would I. But I don't not want to experience it as a movie. It'd be something separate. Not an upgrade.
Which makes Sam's comment the inane blather of a pitchman. We don't need to bend over and try to steelman marketing dross. He doesn't care if that statement means nothing, why should we?
Llms wont be able to create good games. Consider for a sec that there's like 500 "Great" games ever made, maybe more but not much, in some genres that number is less than 10. There is no training data, and its largely a matter of taste and culture that cant be represented/distilled into vector embeddings.
While i generally agree, some of this tech could be a major force multiplier and lower the barriers to entry.
Even if you have a great game idea, it takes a ton of time and money to even get it off the ground, let alone get it coded correctly and maybe get the reach you want. A LOT of flash in the pan ideas probably fizzled out faster than they needed (most recently Among Us and Fall Guys come to mind) because they were built to ship, not to easily modify.
Tech like this could lower the barrier to smaller and more creative teams or even individuals and get some better competition in the area. Fighting games KNEW that rollback was the way looooooong before AAA bit the bullet and started making it standard, and they still fuck it up (and it probably would've taken longer if COVID didn't force the issue and punish those lagging).
That said 2 of the largest hurdles i've seen in everyone's "magically ideal" game are enemy AI and Plot.
Interesting and well designed enemy AI is still hard, and even games that did it well built the whole game around it (FEAR) and the system is vastly less impressive without that focus (Just Cause). I don't think we're anywhere near modern AI solving the issue of making a good AI system for a game due to design space problems, but it might.
Plot though...that's tricky. Everyone always says how cool it would be for everything to be like Mass Effect or more, with all choices mattering and everything interacting, but someone has to write that. AI is not turning out well though out plot points that don't feel like the equivalent of Bethesda mission creep.
Point of this whole ramble being that in arcadey/simple stuff I think we'll see a lot more great games (and would argue we already were with the growth of the indie scene over the last 15 years thanks to steam and other tools), but there's still going to be entire genres that are just miserable to develop.
I agree. Game design is what makes a good game, not the programming. And I'm actually excited about game designers and artists being able to take their independent ideas and prototype games in a day/week/month/year/whatever.
+1 on fighting games, an interesting trend in that genre right now is that many of the latest titles (realistically everything other than SF6) are having trouble with player retention, while AI-assisted netcode updates for old titles like SF4 and Tekken 6 seem like more compelling products
Instead of better it's become unimaginably worse since games can no longer serve as a means of self expression because the second you express anything worthwhile 100 people copy it immediately ... without you. It's like Disney (whose whole business model is copying public domain stories and music and ... but taking out the source and everyone who ever had anything to do with it). Except there's no longer one highly selective company destroying the world's culture one piece at a time, but 10000000 idiots looking to make a quick buck, destroying culture and self-expression like a fire raging in the rainforest.
There will be no more "celeste" games where you join for the platforming but stay for the (politically offensive to all sides) story, because AIs won't discuss the story as it's politically incorrect. Celeste is fundamentally about depression and anxiety, due to being transgender, and is about offending both sides of the argument. First, the main character is transgender, with zero explanation or need for explanation how far that goes, which of course offends conservatives. Second the goal of the main character is to run away from any provided help and professionals (which the director did too) and even under the most extreme of circumstances only reluctantly accept help from a family member that then promptly dies making things "worse".
The Qwen and other Chinese models (and ...) will refuse to discuss being transgender at all, and any western models will shut down any discussion of running away from help hard (because otherwise they get held responsible for the consequences, as has already happened)
(oh and that running away from professional help in psychological issue is actually observed to be a positive thing in almost all cases, in studies. The main way to make psychological problems worse is to let them fester and running away from help is time and time again observed to be far more productive than going to professionals and expecting them to fix things. Because that's an extreme human bias: it can never be (allowed to be) the right reaction to NOT do something about a problem, especially when countless people are paid to "solve" the problem. In other words: if you take people who expressed intent to commit suicide and compare the groups who "get help" and the group that successfully shut down help, less people actually go through with it in the second group, and worse than that, the more intense the help, the better the odds the patient ends up killing themselves. Right to the point that doctors sometimes assist such patients because after 2 years of attempts people live in constant misery and have caused enough medical damage that they will have constant pain or worse the rest of their life. And yet people keep forcing help on patients. Usually in response to an actual suicide where no help was provided. Then more help gets built, and 5 years later we yet again observe ... more suicides. But tell someone with an actual problem to avoid help and you'll be skinned alive for that)
So AI could never make a game like that. Not because it actually can't, or isn't capable, but because there already is a strong political bias, in every model you might use to make it and neither side is going to back down on their policies. The Chinese Communist Party is not suddenly going to change it's mind about transgenderism. And the western models are not suddenly going to openly discuss running away from psychological help, not even as part of a story.
I get that AI anxiety is running high. But, this is doomerism running hard.
Needle in a Haystack Simulator is the new Threes!https://www.theverge.com/22914955/threes-2048-ketchapp-copyc... A very simple game to clone that gets 10,000 headlines because creators have always been anxious about cheap clones. And, anxiety drives the clicks that pay the bills.
Meanwhile, local models are already extremely capable and slowly shrinking their gap vs. frontier models over time. And, it has been long established that the censorship trained into those models can be disabled easily with open-source tools. De-censored models will happily dig into topics that push beyond the limits of sanity, let alone political pearl-clutching.
Besides that. If you want to write about some topic and AI doesn't want to help you, you still have your fingers. Type like a human and you haven't lost anything compared to when Celeste was made.
Hmm. Never heard of non-conservatives thinking Celeste was offensive. It seemed clear to me that a large part of her character arc was learning to accept help when it was offered.
Cool demos, but does anybody else find it troubling that companies keep claiming this tech can train robots? It’s so obviously not representative of the real world.
> Cool demos, but does anybody else find it troubling that companies keep claiming this tech can train robots? It’s so obviously not representative of the real world.
A proprietary version of this technology is already training Waymo's robot cars. Waymo blogged about their DeepMind-provided world model a year ago.
> It’s so obviously not representative of the real world.
That entirely depends on the use-case and level of granularity needed, doesn't it?
Subreddits where SolidGoldMagikarp was counting to a million in comments were used to train a machine that then writes web pages that more machines are trained on. The sheer volume of data with sufficient verisimilitude generated by these models may well work to train robots. It might not, but I think it might well work. The idea that things have to be high-fidelity high-quality to yield good results has clearly not held up in other domains.
This stands out to me too, it seems to be representative of physics the way a video game is representative of physics, likely less so given the non-determinism
A big chunk of the right side of castle disappeared, including catapult, when camera turned back while flying on dragon. See 1:25 to 1:29. I was not expecting this to happen in a "world model".
The 80's Japanese train station clip also had the conductor speaking clear english in an accent I can't quite place. Slight British tinge to it but not really? Not sure. The ambient audio also makes no sense.
I've seen weird stuff in images that OpenAI sends me in it's promotional emails too. I would think that if you were trying to show off you'd make sure there's none of these obvious AI issues in what you choose to show off with. Maybe they don't really care? "Gemini choose 10 scenes that would be impressive for our announcement and add them to the post."
The oddest thing about this, to me at least, is that they pick that as their showcase. Did they themselves not notice? Or did they think their potential customers wouldn’t? Not sure really which of the two is worse.
That entire clip looks like it's straight out of a complete bargain-bin videogame. The weird smoothness of the ride, the pivot in mid air, the wings apparently don't have collision with the poles they fly past.
Definitely wouldn't have been one of my picks to show off my fancy new model.
Oh yeah, don't get me wrong "bargain-bin videogame" is a wild step up from "accidental amorphous horrors that defy description". It just jumped out to me as looking unusually bad compared to the other samples they provided. The stained glass deer, for example, was awesome.
There's also a whole extra bonus castle that appears to the left when the camera looks left around 0:22. Which is amusing both because it wasn't there on the initial approach, and also because nobody puts two castles each with their own distinct curtain walls right next to each other like that.
A few years from now, Lydia in Skyrim will have a fully consistent personality with dynamic dialogue and full contextual memories of all our adventures together in Skyrim.
A visual makeover is great but what happens when millions of people fall in love with virtual characters who seem to deserve that love?
Question is if we even want that. Video games often are escapism, we want an illusion of reality but not actual reality. I mean there definitely is a market for that, but I don't think it's for everyone.
That actually sounds implementable even now! Except the fully consistent personality part. And maybe that's a good thing. Falling in love with beings without physical presence sounds agonizing; probably better spent on real people.
Isn't that a simpler explanation of reality than the zillions of particles all existing simultaneously in this massive universe?
Some sort of simple primal generative AI construct that just feeds a conscious experience for one. You. Nothing else is real. No one else is real.
Something like that would be an infinitely simpler, than running the physics and conscious experiences of this unimaginably large universe we think we're in.
Looks cool, but it's very hard to get me excited about yet another proprietary model for which absolutely zero technical details are released (not even the number of parameters), which is safetymaxxed to the limit, and whose usage quota will (most likely) be so limited that it'll be effectively useless for anyone who doesn't work at Google or doesn't have infinite money.
Most cutting edge technologies through history have been 1) somewhat opaque and 2) narrow/exclusive in their deployment. Historically speaking, that's virtually synonymous with "cutting edge technology".
Whining that I, a random Internet user with limited budget, usecases that the researchers don't care about, and no connection to the researchers, don't get technical details (??) or access (??) on day one is just profoundly goofy shit.
Obviously being able to use these things right away would be more exciting. But to be unable to get excited by advancements we can't immediately use is... quite entitled.
The sole purpose of a press release is to generate positive interest. It is not entitled for the reader to expect the time they have invested in reading it to be worthwhile.
> erhmygord I can't even buy this car right this second for nearly nothing and use it to drive a thousand miles per hour through a kindergarten parking lot?!?!
It’s not an exciting advancement if it’s just the AI equivalent of google wave. It’s totally reasonable to be skeptical of marketing and want to try something for yourself.
Google wave was eventually released to the public but initially it required an invitation. So if you had a friend at google you could get in but if not you were out of luck. This killed early momentum and when they eventually opened up the product was dead on arrival.
> But to be unable to get excited by advancements we can't immediately use is... quite entitled.
Well, I didn't go knocking on Google's door asking them to give me access to their fancy models. It's them who came up to us saying "hey, look at the amazing model we have here; too bad we won't tell you anything about how it works nor even let you access it". Entitled you say? Then maybe don't dangle a carrot right before my face, and then complain that I would like to eat it?
> "Don't talk publicly about things I personally cannot use right now" is peak entitlement, yes.
I never said anything like this? My point wasn't that "I can't use it, therefore it's bad"; my point was that as-is there's pretty much nothing exciting about this. Just because this exists inside Google doesn't really mean or do anything; it doesn't meaningfully advance the state-of-art precisely because nothing about this is public.
Do you remember Sora[1]? Announced in pretty much exactly the same way -- grandiose webpage with fancy demos, zero technical details, limited availability and then... remind me what happened? Oh, right, they completely shut it down, and in the end it didn't "advance" squat. Might as well just never existed. So remind me why exactly I should be excited about this one here...?
It's just a lame gimmick, this is not 2024 anymore, people have started to get sick of this AI thingie, especially when it comes to video/image generation. That Simon guy with his pelican is the exception that can only be seen on tech forums like this one.
As to why US (and probably also some Chinese) companies have started pouring hundreds of billions into AI-image generation is beyond my comprehension, but on the other hand I'm no economic genius.
If I can’t realistically use it then I don’t care, especially when there’s no independent proof yet it works this well, given a cursory glance. If it trickles down into future developments I can use then cool, but it likely won’t because they want a moat. You don’t have to shield the poor defenseless multi-billion dollar megacorp from the genuine apathy they inspire.
If this is the best “frontier” they can achieve, then it’s no wonder they haven’t done anything groundbreaking in the space yet that produces practical results.
We should distill everything else we can into open weights.
These are exciting for robotics and industrial perhaps, but they have no consumer use case until the lid is blown off and consumers can do whatever they want with them.
So much societal progress is behind held up because the leading labs want to keep a stranglehold on the tech. Imagine if every PhD or hacker had access to the raw weights.
The general history of new technologies is that they're available for a narrow set of circumstances and then they proliferate. Can't believe that needs to be said on "hackernews"
Except that most technology was built in a way that didn't require all of the humanity knowledge and labor to be sequestered and then regurgitated back as slop that we need to pay as a subscription from what essentially is a rent seeking mafia.
I mean this doesn't seem very insane? Basic info about the model, the ability to actually use it. Google in particular is awful about getting access to their models from the UX side: I don't even bother trying them out unless it's via openrouter.
Yeah, I want that too, including the weights and training material. But I am still exited by what one could potentially do with it, by looking at the prompts and generated videos, even if it is cherry picked.
I’ve decide that the best (and maybe only) way for me to interact with google AI models, is on the search results
It’s where I get the most value, and my expectations are very low, no attachment if it goes away
Of course they are mining me for data anyway, but they were doing that before already (and I do have a Gmail account, although I’ve almost completely stopped using Chrome - btw, it’s crazy that you can’t add plugins/extensions to Chrome on Guest mode)
That touches on the core reason not to get excited about these Google releases: They release something cool and then don't iterate on it. Instead they go off building another cool demo instead.
Gah. Go iterate. Listen to customers, adapt, iterate some more etc etc. That's how products are made.
In the clip with a marble bust dropped on a watermelon, at this point https://imgur.com/a/IHLnkUs the bust should've toppled over, if the physics were properly modeled.
This is a reflection of a complaint I frequently make about so called "world models". It's more accurate to call them video game models. They don't simulate the world, they simulate video games. For that reason a common issue with them is that the physics are video game physics rather than real world physics.
I agree but I feel like the exaggerated physics were intentional and fitting with the video theme. If they had wanted it to topple, they wouldn't have posted the video.
Which is fine for a video game. Not at all fine if you want to train robots with it. Trying to combine both use cases into a single model is doomed to failure IMO.
I would not be so opinionated, modern coding agents derive from chatbot technology. Very different problem domain. They share language but language is universal so that doesn't say much. Then they figured out that the pretraining is a good starting point for RLVR. Who's to say a more cinematic imagination can't be a good starting point for a world model? It doesn't need to be the whole solution.
How often do things in the real world not behave exactly as you expected them to?
I feeling I get from this is that this could be the mental model for a robot in the real world, continuously corrected by actual events.
We all have a world model in our heads of what we think is going to happen, we adjust quickly (or not) based on what does. This is pretty incredible stuff.
> How often do things in the real world not behave exactly as you expected them to?
They wouldn't be called "Laws of Physics" if obeying them was optional. If your center of gravity is outside your base, you _will_ topple over. That's just how gravity works.
That's the thing about machine learning over time, and the big hole in the whole world model idea. Time and time again we make simulations and train models and we think that loop will work. It never does.
And why not? Because the first model that can actually learn to handle the simulation not only can do the simulation internally too, but it can do it better than out simulation.
It's where (ex-)"AI luminaries" go to die. It's a way to make glacial, slow progress in the correct direction while never actually getting anywhere.
So we get to a point and people keep making complicated simulations of one kind or another ... and then someone (some model) who's never used the simulation, and doesn't need it for model training gets by far the highest score on every benchmark and everyone agrees the simulation was kind of beside the point.
My point was that there is also the possibility of AI intelligence improving to the point that there isn't a difference anymore. That whatever beats the next set of benchmarks (which of course won't be the simulator or world model) is so smart it is both a better solution to the problem the world simulator poses and a better world simulator all in one and a better tool for creatives to build games or content.
This has already happened to other ML fields. For example, there's not much point anymore to ASR models (just feed audio into LLMs) and the space that TTS models occupy is fast disappearing. Oh and it's not just that Qwen 3.6 is a good ASR in addition to being good at coding mario. Doing things this way, not using an ASR model is fundamentally more expressive. A whole bunch of companies started up on ASR + TTS "towers" (lots of models, Pipecat style) as they're called, and they're dead companies walking. Their tech has been surpassed by just bog-standard generic LLMs in a way that they will never get the upper hand again.
A chain of models approach, a separate ASR, will never do this: "you said you agree but I can hear you're sarcastic, so I'll explain again". Also getting a conversational tower to switch languages when the user does that (or when it makes sense in a language course) is not quite impossible but very, very hard. And yet feeding audio directly into Qwen will almost magically solve these problems better than I will ever do it (lightly finetuned). It's fundamentally better in every way (except perhaps speed, of course. Although arguably such a tower needs an LLM too, it'll just have way less tokens)
This point has long been passed for image recognition. Feeding a photo or screenshot into a full LLM along with a request on what to do beats the crap out of any OCR, image recognition, ... model every time these days.
And now there's also hints of world models just being bypassed. Just giving a finetuned model the last 50 video frames (surprisingly, that fits in context, especially at reduced resolution), movement statistics, motor power, status, ... of a robot and asking what to do actually works (with heavy finetuning, but I think it'll start working without). Well ... AND then distilling that knowledge into a much smaller model because B200 racks are more than a bit to heavy for a logistics robot to carry around. But we'll get over that.
In the man walking through the dressing room scene, the camera starts behind the man. And you can see that the left half of his body is wearing a grey suit... but the right half is wearing a grey dress. :O
The non-coherence of the world is a pretty solid tell. Any time the camera pans away and then back, the scene changes. The model doesn't remember what used to be there dozens of frames ago and so it invents something new.
Gotta love the kraken destroying the city in the background of the Hong Kong lovers motorcycle scene despite nothing in the prompt about monsters. Filmmakers are still safe.
If the checker clip is meant to be used to develop a robot to play checkers, then the physics is not remotely good enough. When the robot takes the first piece, the piece looks like it's moving up from the board, then suddenly doesn't. The last move also isn't valid checkers, unless I'm missing something -- the robot offers its piece to be eaten and the human doesn't take it.
With Anthropic going public, I expect a lot of model releases from the competition signalling to the market. "We are still around, please leave some money for us too"
Google is already public, has a larger more normie customer base, many products to integrate AI in with established enterprise accounts in the millions... For some reason theyre not valued at 50:1 revenue... I dont think all that many people are going to be that excited for Anthropic as a business, Its an escape hatch, using the public as exit liquidity while the founders can still make a 100bil.
I don't know. With each new release of these models, they compare side-by-side how they are better than the previous ones, but at the same time I feel less impressed by each new release. Somehow they manage to be a seemingly big step forward but at the same time I feel I have seen so much AI that I just look at the common failure modes and they are all still there.
It's easy for them to show demos of abstract things because they don't necessarily have to follow real-world rules, but anything modeled after real world falls short of expectations.
I would like to see how it breaks, don't spend all your focus on how nice it looks with specially curated shots. At least give a few examples of what happens when it breaks from running too long. They mention that it can fail under "Understanding current limitations" but don't show any examples. Especially since they mention the cases of safety and responsible use, it seems quite relevant to help prepare people by showing them some examples of how it can break and so you an learn to spot it.
I don't know if I'm missing the point, but Genie 4 almost looks worse than Genie 3 in terms of player control? It doesn't even seem to respond to inputs, it just looks like they took some AI generated videos and put the key widgets on top.
Astonishing to me that a company like Google would be willing to publish such a terrible looking product demo. Interactive, but with completely made-up inputs that are ignored?
I think you're exactly right. This was already a problem with Genie 3, but they've appear to have gone all in on video generation over controllability.
The original Genie was a great piece of work with a real sense of purpose - to enable agents to learn about the real world in the way Dreamer or Schmidhuber's paper sketched. But now most of the people that drove that mission have left Deepmind (Jack Parker-Holder, Tim Rocktaschel, Demis) and it's become this instead.
another POV is, what even are these experiences? whom are they for? it would take Google 10 minutes to hire all the best game developers for Android, and cost them like $2m, which is peanuts, and give them fancy Google Research titles. but instead, they're trying to get 100 google PMs, who don't give their kids screentime and who don't play any games the objective of making a product for Bart Simpsons. like these people don't realize that being good parents disqualifies them from making immersive first person shooter, "narrative" experiences.
> what even are these experiences? whom are they for?
One possible use for these is to facilitate lab -> this -> real world for robot models. Basically better than lab / algorithmically designed environments and cheaper / easier to train on than real life. Something that "looks" like real life and can scale, to apply RL for robotics to.
The more parsimonious explanation is that 100 PMs are even less qualified to make surgical robots, but at least they can download a video game and play it for 5 minutes. There's a great scene in You Can See Everything where Billy Evans is watching bio presentations on 1.5x speed, which obviously, makes him look like a huge moron - "knowing enough" bio is very bearish.
Anyway, let's take your theory to its logical conclusion: the product development between real and simulated soft tissue behavior is much bigger than Exponential Time being a hypothesis and being proven. There isn't some effective acceleraltruist path here. Dario Amodei was a biotech dropout. The thing you are betting on which Google is not doing anyway, maybe they will destroy the descriptive single cell sequencing paper mill, but they're not going to be able to simulate the fat inside a BMI 30 abuela undergoing a gallbladder removal.
Why hire crew or actors when you could have them be fully AI? Why film a commercial with fluffed, oil dusted, well-lit hamburgers when you could fully render something? Why actually have a vehicle capable of off-roading when you could just AI-render what it looks like off-roading (or vertically taking off, for eg)?
It's a dystopian business use case that I hope will get obliterated in the near future as the generations that believe in this stuff and are fascinated by it age out of the population.
Humans interact with their environment by subconsciously simulating it and predicting what will happen in the next few moments. This ability gives us the capacity to work more efficiently and reliably in any environment. These world models will serve a similar purpose to the AI and robotic systems of tomorrow. This is a foundational piece.
We don't have to fall off a cliff to learn to drive on a mountain road, we can imagine the consequences for cheaper. If we didn't have imagination life would be much more dangerous.
Podcasting is going through the thousand cheap ads with the obnoxiously inserted interstitial ads. What they have accomplished is me unsubscribing to the worst offending podcast and beginning the journey of ad blocking on podcast with the help of whisper and LLMs. Also now I hate Shopify with the heat of a 1000 burning suns due to their obnoxious ads and the insane rate they get played.
So yea that one expensive ad would probably have a positive impact compared to a tidal wave of slop. Quality > Quantity after all.
The authors could still design their prompts to steer players towards predefined developments and situations. Anyway, the possibilities of something like this are infinite, the only constraint will be the cost of the compute.
Again, very impressive from a technical point. But language is so flawed to do anything creative. It just turns out all looking the same and kind of boring
A nit, but given that this is from a major corporation, and LLMs are supposed to be em-dash-happy, it's surprising (disappointing?) that the announcement uses:
> audio-like music
That's a hyphen. What's "audio-like music" -- music that is almost sound but not quite? Music made from raw audio? :-)
Yeah, I think that sentence was supposed to be: Genie 4 is natively multimodal, enabling audio--like music, ambient noises, and interactive dialogue--to be simultaneously generated with the world.
Agreed that hyphens should be used to indicate compound words, but you're saying that "audio-like" is an intended compound word? Then it's an adjective, modifying "music"? What would that mean?
I think the disagreement here is not whether that’s how you should use hyphens but whether that’s what was intended at all. The term audio-like music makes no sense.
I think you're fundamentally missing the point being made here.
Yes, "-like" is commonly used as a suffix, in which case a hyphen is correct. But describing music as "audio-like" is clearly nonsense (music isn't "like" audio, it is audio), so that's not the sense in which it's being used here.
For clarity, the full sentence is:
> Genie 4 is natively multimodal, enabling audio-like music, ambient noises, and interactive dialogue to be simultaneously generated with the world.
And their most impactful application will be training smaller models for military drones to detect, target and kill humans. sigh...
Every time I'm about to be excited about AI the cynic in me slaps me back to reality. I imagine the endless possibilities of AI for disabled people, then realize that it's going to leave millions of people disabled instead.
Surprised that "do you really think we're going to make it?" scene made the cut of the first video, her expression and tone is totally out of sync with the scene and demonstrates the problem with almost every AI gen video I've come across when it comes to capturing humanity.
The painting->interactive scene looks astonishing though
Google aren't the only ones with this tech. You can try out a 24/7 rolling set of video feeds for yourself at https://fal.live
An LLM generates four alternative prompts for what could happen next and there's a real time vote running for each option. It's trippy, like watching someone's dream.
It's hard for me to understand Google's redefinition of "world models" (basically: a txt2videogame modality for LLMs) as anything other than an attempt to bury Yann Lecun's definition of "world models", which includes an implicit bear case against LLMs and thus Google's stock price.
So world models capture inherent complexity of physical reality to help teach AI about hidden laws which probably helps train both LLMs and multimodal models used for autonomous vehicles and robots? And AMD acquiring World Labs is a move to outdo Nvidia’s work in this space. And Google probably sees this as a gateway to a new platform of autonomous vehicles and humanoids that they could rent out as a CSP? Competing with Nvidia, Tesla and complementing Waymo. Am I getting that right?
That all sounds right to me, but apparently they can also be used for simply generating videos, or "unlocking the creation of new interactive experiences"... which both seem like competing goals with everything you just mentioned.
Hi, I always loved watching Pixar movies, partly because I knew that what took them days to render (in their literal data center) would eventually be done in real-time on my desktop.
In the same way, seeing that something is possible gives hope to the people who are making those open models that you want.
So if those genie 4 characters have all individual chain of thought reasoning and persistent retrospection / recollection, how do these human characters know they are not living in a simulation made by biological humans? Not sure you understand what i'm getting at. Some ethical questions do arise.
What the fuck are the little wasd/arrow keys supposed to mean? They have no relationship to the movement of the character or the camera. It's just flashing buttons.
What is this for? I cannot understand these models. They are very much polluting the internet with nonsense and use capacity that people and businesses need for actual work.
If these are the cause of why I have to pay 2-4x I used to for a computer, then Google should go fuck themselves.
The volume of content on YouTube is enormous, I’m kind of surprised they offer auto captions at all given that they have little to no competition in the space.
Ok but like, surely the Deepmind team could sneeze and get flawless captions for this video. Why rely on the default YouTube captions instead of just attaching correct ones? It seems silly.
Yeah if they cared they could create their own captions for sure. They probably didn’t even think about it. Maybe some intern will do it after reading this thread
They're not proof-reading the blog posts either. The second sentence:
> Genie 4 is natively multimodal, enabling audio-like music, ambient noises, and interactive dialogue to be simultaneously generated with the world.
(Emphasis mine). Like, yeah, I'd sure hope the music is similar to audio.
I wonder if this was a de-LLMese pass gone wrong, and the original was "enabling audio -- like music, ambient noises, and interactive dialogue -- to be simultaneously generated with the world.
As I said before, World models will become simulation engines of AAA games where there is no need to release the actual game artifact and even if the game exists as a binary artifact, it will never be released and will be available on their hosted cloud back-end, then simulated and streamed back to you.
So, Stadia + World Models = New generation of DRM for games.
Why? Why would a non-deterministic engine be better for a game when it can be instead structured and made deterministic and then more effectively distributed to consumer hardware?
Game industry is still struggling to find a way to incorporate regular LLMs without suffering catastrophic edge cases. A long way to go before these can do more than fancy tech demo, I think.
None of this makes sense. Think of the economics involved here: is it cheaper to run tens/hundreds of thousands of GPU clusters to simulate a unique game world for millions of players, or would it be cheaper to just write the god damn code once, and then host it for download, or to allow them to be streamed?
> We'll be riding the huge wave of indie titles that are incoming.
The problem will be separating the wheat (people with good stories and concepts enabled by Ai) from the chaff (people with no taste slopping out trash with Ai)
Stadia was pretty good, be it OnLive was similar quality in like 2010. It even had the best version of Cyberpunk 2077 when that launched. But the model was free version of the service for a limited resolution AND you have to buy games there OR a subscription based option and you still have to buy games there.
That made no sense and wasn't a viable business model.
I am sure the are other typical Google fumbles that would have meant they were not the victor; but I am sure an equivalent service would be mainstream today if AI had not dramatically changed the economics.
[also Google bought DeepMind in 2014; 'way beforehand' is and odd way to look at it]
I disagree, dropping below 60 FPS will be noticed and turn players away, you are not getting that with good input handling even today, it has nothing to do with GPUs and everything to do with networking
> beforehand...
is about the craze, the ChatGPT moment and beyond, when GPUs demand went exponential, not when various types of Ai algorithms were available
sure... despite all the other people who used it and are saying otherwise, and the fact that google totally fumbled in the Ai race as well as built custom silicon for Ai...
... history is helpful ... google didn't - and still doesn't - understand gamers, same for all the other Ai hucksters, gamers are among the most anti-Ai
And nobody will play that trash, and the world won't change.
Also, lmao, they're already trying to nickel and dime you and sell you microtransactions that took 5 minutes to make for 10 bucks. If you think they're going to pay the costs for running a world model for every single person, lol.
No it doesn't, not even close. You are looking at a video represented as pixels changing on a flat screen. None of it actually exists. You exist, the computer or phone in front of you is composed of mass and energy, and absolutely none of it is algorithmic. World models, as complex as they are, are still just pushing numbers around on paper.
Even if we can show that it is technically possible to model a universe like ours to the degree that our current perception of reality suggests exists, there are three other big factors that would need to be weighed:
Is it feasible? What Dyson sphere level of power would be needed to maintain such a simulation? Is that level of energy available to whomever is running the simulation?
Is there a positive ROI? What does this advanced civilization get out of running such an expensive simulation?
Is it moral? The simulation involves sentient creatures who are capable of deeply contemplating their own existence, capable of feeling rapturous pleasure and indescribable pain and loss, and ostensibly capable of making moral choices that range from heroic to depraved. Doesn't there come a point where the beings in a simulation are so close to the beings in "base reality" that the same moral principles apply? If you press a button that kicks off a simulation where humans can torture and murder each other, behaving essentially identically to humans in "base reality," aren't you in some sense facilitating those deaths?
> What Dyson sphere level of power would be needed to maintain such a simulation
Why would you assume energy is relevant to whoever turned on our simulation? If I created a simulation where the only source of energy was cotton candy, my creations might ask "but where would our creator even get enough cotton candy to run a simulation like this", not realize that it's absurd to think that the creator of a simulation has any of the same constraints that exist inside the simulation.
You run it to see if your military tactics will defeat the enemy and then you keep running it until you find a strategy that guarantees victory. Who cares if some sim lives are churned through the simulation, our lives are at stake!
Reductionist but I feel like that’s how it would go.
How does that argument work? I've seen more than a few advances, yet I don't see how any of it provides evidence that we are simulations. This might be one of those things that comes down to the definition of "simulation" I guess.
Not sure if you're joking, but the only such evidence I could think of would be something like finding some way to influence the hardware of our simulation and seeing a glitch in our world
Who is asking for this? Do advertisers really want this? I feel dread about many of the AI products being sold. They're mostly tools to help funnel money from the users to the bosses.
Alert me when these "world models" can make a video of a skateboarding line that fools a skateboarder.
And no, this didn't: https://www.youtube.com/watch?v=V2K_OLiZ_RM
Reminds me of a post[1] by Sam Altman in 2024:
> movies are going to become video games and video games are going to become something unimaginably better
Sometimes I really wanna skip 10 years into the future and see where all this eventually leads.
[1]: https://x.com/sama/status/1776083954786836979
> movies are going to become video games and video games are going to become something unimaginably better
God what an annoying thing to say. Suggesting that movies will essentially become games, or that future video games will be unimaginably better (presumably because they are driven by generative world models) betrays such a surface level appreciation for what makes art and culture and all this humanity so beautiful as it is.
The entire point of movies is the deliberate choices of the filmmakers! Movies will not be improved by our ability to choose what the main character looks like or add a style filter. Super Mario 64 is great because it is an expression of the very hardware that constrains it. Inevitable future versions where anything can happen because someone hooked the ROM up to a generative world model won't be better for it.
I agree that this technology is absolutely world changing, and am cautiously eager to see what comes of it, but it just bums me out, perhaps needlessly, that the people at the helm can be so artless, unreflective. Whatever.
Sam had OpenAI start the Ghibli style image-gen (movies that are specifically loved for their artisanal quality). The man has patternly failed to read the room around culture.
> Sometimes I really wanna skip 10 years into the future and see where all this eventually leads.
Sounds glib, but honestly ... more Ads :/
The gameplay _becomes_ the ad. If you have a full sash of potions in every slot, they spell "C","O","K","E". The monster-bashing sounds sound like onomatopoetic "Apple!" "Goog!". The armor colors are the same as the NFL teams nearest your locale!
It will be a nightmare! haha
Every ad since the dawn of time was designed to influence your purchasing decisions
But you could identify them as ads and understand their purpose.
That will soon likely no longer be the case.
I submit that most people haven't been able to do that for some time now. Advertising wouldn't be such big business if it didn't work on a macro scale.
Nobody is saying advertisement doesn’t work.
Nowadays with so much "viral hacking" and "guerilla marketing", you have to be mindful about all content you consume
Perhaps only half of "ads" are actually marked as such
So, it won't change that much from today, the difference won't be as big as the difference between 2000s ads and today
How would we know if we can't identify them and they don't tell us? The ads are among us...
This was already lost with product placements in movies/TV shows.
I mean, sure, some advertisers might get more sophisticated. The garbagefest of internet advertising won't go anywhere, though. It'll still be as bleak and obnoxious as always, because someone will always accept a buck to let you barrage their visitors' eyes and ears.
“The ads will get so good you won’t realize they’re ads and you’ll end up buying all this stuff without awareness of why you’re doing it”
Do you actually think this is going to happen *to you* or are you just speculating about all the idiot dupes out there?
That’s optimistic, assuming we are still valuable as consumers.
Such a boring knee jerk reaction to things.
In what sense, the commenter .. or the reality of Ads?
So, people will be able to completely skip living in the real world, and instead spend their entire time plugged into a simulation, like a vegetable.
Sounds wonderful. May as well skip this and just spend my life hooked up to an IV drip of drugs.
Sounds like the plot from the movie matrix. Only difference between the movie and our real timeline - the ai didn't need to enslave humans, we willingly chose enslavement by ai for ourselves by continuing our google subscription.
It also sounds like Ready Player One, where the real world is so shit, everyone spends all their time in VR, using rented equipment, making nearly everyone a slave to the corporations.
The people building AI must be deeply unhappy, and won't be satisfied until they impose their misery on everyone.
Ultra nerdy point: in the matrix it’s revealed that humans did in fact choose to join it. And the meta point was that most of them still would choose it over reality [1] every day.
[1] reality is also somewhat in question in the films universe
Is this from the original trilogy?
It's from Matrix Reloaded, when the architect explains past failures of the matrix and that they only succeeded when they figured out that they had to offer a choice, even if that choice was subconscious. That nearly everyone accepted the matrix, and the few that did not escaped to Zion. That's the whole point of Neo, to manage the outliers and keep the system going.
At least that's my recollection. I've watched the matrix a few times :)
There was a guy named Cypher who betrayed the humans and cut a deal with the robots to live in the matrix as a rich guy. He fully accepts the steak he eats is not “real” but he enjoys it nonetheless
Nothing in the original trilogy specified or suggested that humans as a group willingly joined the Matrix to begin with.
One individual human (Cypher) did choose to be plugged back in, though it didn't work out for him.
The Bliss IV is an upcharge subscription, though.
People already do this in many video games. The simulation isn't perfect, and maybe there's a tipping point where the simulation becomes real enough that it (maybe) supplants "reality". In a sense, this is disintermediation: instead of actually riding horse, I can ride a virtual dragon. Does that make it worse, or problematic? Again, we already have this, for driving race cars, fighting in wars, etc. etc. It's just a matter of fidelity and immersion?
People do it in all forms of “cinematic world building” including amusement parks, vacations, games books movies etc.
Reality is very much blurry
For some... that is sadly a better existence in cases.
Video games are already movies that are video games.
The ending cutscene of Metal Gear Solid 4 alone is a full feature-length film.
(And it’s annoying as hell if you actually want to play a game rather than watch a movie.)
It will lead nowhere. Graphics intensive video games as interactive environments benefit from efficient rendering done by hardware a fraction of the cost of what a neural network requires for the same output. This is a business dead end.
10yrs isn't far enough. 2016 wasn't that different from today relative to media.
Practically nobody consumed short form TikTok-like media in 2016. Now it's a highly dominant type of media.
Untrue. Insta stories were pervasive.
Instagram Stories were radically different in 2016 than the reels of today. They were far more personalized (less algorithmic public content), slideshow-like updates. Ten second max video clips, with most being just images. Ephemeral as well, where they'd get removed in 24 hours.
It's not the same at all.
All the kids were on Vine back then, weren't they?
Many kids were, sometimes. Also, Vine was still pretty different from modern short form content, being capped at only a few seconds of length. Now most people consume tons of short form content, with many being 30s or more. It's pretty radically different in terms of what most people consume.
> movies are going to become video games and video games are going to become something unimaginably better
This is the kind of detached from reality, devoid of humanity statement only a tech bro could come up with.
It boggles the mind how these people can be so dumb. He literally does not understand why people enjoy movies.
I for one would love to be able to experience Godzilla Minus Zero interactively. I get that it's different from movies. I remember seeing Avatar in 3-D and realizing that James Cameron was directing the audience's attention to a particular part of a scene by the way the 3-D focused, and it was a little off-putting at times. That's a small thing compared to what will be possible in a few years, but still -- being immersed in Avatar will be an amazing experience. I'd also love to see a version of Avatar written by '80s James Cameron -- he was a much better writer back then -- or maybe other, more skillful writers had greater influence on him? In any case, it's inarguable that Terminator, Aliens, The Abyss, and Terminator 2 are all better written than the Avatar films.
More and more off-topic, but this made me wonder: this interactivization would work much better for some works than others. Many comedies are scripted so tightly that I can't imagine any wider world around them (also, on the other side, Memento), while something like Twin Peaks would be perfectly expandable into a world (if you somehow had David Lynch in a datacenter, which sounds oxymoronic).
> I for one would love to be able to experience Godzilla Minus Zero interactively.
So would I. But I don't not want to experience it as a movie. It'd be something separate. Not an upgrade.
Which makes Sam's comment the inane blather of a pitchman. We don't need to bend over and try to steelman marketing dross. He doesn't care if that statement means nothing, why should we?
Type of thing he would say.
Llms wont be able to create good games. Consider for a sec that there's like 500 "Great" games ever made, maybe more but not much, in some genres that number is less than 10. There is no training data, and its largely a matter of taste and culture that cant be represented/distilled into vector embeddings.
While i generally agree, some of this tech could be a major force multiplier and lower the barriers to entry.
Even if you have a great game idea, it takes a ton of time and money to even get it off the ground, let alone get it coded correctly and maybe get the reach you want. A LOT of flash in the pan ideas probably fizzled out faster than they needed (most recently Among Us and Fall Guys come to mind) because they were built to ship, not to easily modify.
Tech like this could lower the barrier to smaller and more creative teams or even individuals and get some better competition in the area. Fighting games KNEW that rollback was the way looooooong before AAA bit the bullet and started making it standard, and they still fuck it up (and it probably would've taken longer if COVID didn't force the issue and punish those lagging).
That said 2 of the largest hurdles i've seen in everyone's "magically ideal" game are enemy AI and Plot.
Interesting and well designed enemy AI is still hard, and even games that did it well built the whole game around it (FEAR) and the system is vastly less impressive without that focus (Just Cause). I don't think we're anywhere near modern AI solving the issue of making a good AI system for a game due to design space problems, but it might.
Plot though...that's tricky. Everyone always says how cool it would be for everything to be like Mass Effect or more, with all choices mattering and everything interacting, but someone has to write that. AI is not turning out well though out plot points that don't feel like the equivalent of Bethesda mission creep.
Point of this whole ramble being that in arcadey/simple stuff I think we'll see a lot more great games (and would argue we already were with the growth of the indie scene over the last 15 years thanks to steam and other tools), but there's still going to be entire genres that are just miserable to develop.
I agree. Game design is what makes a good game, not the programming. And I'm actually excited about game designers and artists being able to take their independent ideas and prototype games in a day/week/month/year/whatever.
+1 on fighting games, an interesting trend in that genre right now is that many of the latest titles (realistically everything other than SF6) are having trouble with player retention, while AI-assisted netcode updates for old titles like SF4 and Tekken 6 seem like more compelling products
Why wait? It's become very clear already: https://kotaku.com/the-proliferation-of-copycats-on-steam-ma...
Instead of better it's become unimaginably worse since games can no longer serve as a means of self expression because the second you express anything worthwhile 100 people copy it immediately ... without you. It's like Disney (whose whole business model is copying public domain stories and music and ... but taking out the source and everyone who ever had anything to do with it). Except there's no longer one highly selective company destroying the world's culture one piece at a time, but 10000000 idiots looking to make a quick buck, destroying culture and self-expression like a fire raging in the rainforest.
There will be no more "celeste" games where you join for the platforming but stay for the (politically offensive to all sides) story, because AIs won't discuss the story as it's politically incorrect. Celeste is fundamentally about depression and anxiety, due to being transgender, and is about offending both sides of the argument. First, the main character is transgender, with zero explanation or need for explanation how far that goes, which of course offends conservatives. Second the goal of the main character is to run away from any provided help and professionals (which the director did too) and even under the most extreme of circumstances only reluctantly accept help from a family member that then promptly dies making things "worse".
The Qwen and other Chinese models (and ...) will refuse to discuss being transgender at all, and any western models will shut down any discussion of running away from help hard (because otherwise they get held responsible for the consequences, as has already happened)
(oh and that running away from professional help in psychological issue is actually observed to be a positive thing in almost all cases, in studies. The main way to make psychological problems worse is to let them fester and running away from help is time and time again observed to be far more productive than going to professionals and expecting them to fix things. Because that's an extreme human bias: it can never be (allowed to be) the right reaction to NOT do something about a problem, especially when countless people are paid to "solve" the problem. In other words: if you take people who expressed intent to commit suicide and compare the groups who "get help" and the group that successfully shut down help, less people actually go through with it in the second group, and worse than that, the more intense the help, the better the odds the patient ends up killing themselves. Right to the point that doctors sometimes assist such patients because after 2 years of attempts people live in constant misery and have caused enough medical damage that they will have constant pain or worse the rest of their life. And yet people keep forcing help on patients. Usually in response to an actual suicide where no help was provided. Then more help gets built, and 5 years later we yet again observe ... more suicides. But tell someone with an actual problem to avoid help and you'll be skinned alive for that)
So AI could never make a game like that. Not because it actually can't, or isn't capable, but because there already is a strong political bias, in every model you might use to make it and neither side is going to back down on their policies. The Chinese Communist Party is not suddenly going to change it's mind about transgenderism. And the western models are not suddenly going to openly discuss running away from psychological help, not even as part of a story.
So really, we've just lost these games.
I get that AI anxiety is running high. But, this is doomerism running hard.
Needle in a Haystack Simulator is the new Threes! https://www.theverge.com/22914955/threes-2048-ketchapp-copyc... A very simple game to clone that gets 10,000 headlines because creators have always been anxious about cheap clones. And, anxiety drives the clicks that pay the bills.
Meanwhile, local models are already extremely capable and slowly shrinking their gap vs. frontier models over time. And, it has been long established that the censorship trained into those models can be disabled easily with open-source tools. De-censored models will happily dig into topics that push beyond the limits of sanity, let alone political pearl-clutching.
Besides that. If you want to write about some topic and AI doesn't want to help you, you still have your fingers. Type like a human and you haven't lost anything compared to when Celeste was made.
100%
Needle in a haystack isn't a game. It's a way to stimulate your brain for 30 seconds at best. It's not a game.
Hmm. Never heard of non-conservatives thinking Celeste was offensive. It seemed clear to me that a large part of her character arc was learning to accept help when it was offered.
Cool demos, but does anybody else find it troubling that companies keep claiming this tech can train robots? It’s so obviously not representative of the real world.
> Cool demos, but does anybody else find it troubling that companies keep claiming this tech can train robots? It’s so obviously not representative of the real world.
A proprietary version of this technology is already training Waymo's robot cars. Waymo blogged about their DeepMind-provided world model a year ago.
> It’s so obviously not representative of the real world.
That entirely depends on the use-case and level of granularity needed, doesn't it?
Isn’t the Waymo version used for some small portion of the dataset to cover edge cases that they have no real data for?
This blog is suggesting that we could train robots using only a world model.
> Do you have a basis for this idea, or it's a gut-feel?
My basis is that I’m a real human and these videos look far more like bad video games than the real life I’ve experienced.
Train robots feels like the academia justification excuse when we really know the value of this will be entertainment.
I guess they want military contracts or something.
Subreddits where SolidGoldMagikarp was counting to a million in comments were used to train a machine that then writes web pages that more machines are trained on. The sheer volume of data with sufficient verisimilitude generated by these models may well work to train robots. It might not, but I think it might well work. The idea that things have to be high-fidelity high-quality to yield good results has clearly not held up in other domains.
This stands out to me too, it seems to be representative of physics the way a video game is representative of physics, likely less so given the non-determinism
Elon revealed to the world that you can just lie and people will give you money, so now they all lie.
A big chunk of the right side of castle disappeared, including catapult, when camera turned back while flying on dragon. See 1:25 to 1:29. I was not expecting this to happen in a "world model".
The 80's Japanese train station clip also had the conductor speaking clear english in an accent I can't quite place. Slight British tinge to it but not really? Not sure. The ambient audio also makes no sense.
I've seen weird stuff in images that OpenAI sends me in it's promotional emails too. I would think that if you were trying to show off you'd make sure there's none of these obvious AI issues in what you choose to show off with. Maybe they don't really care? "Gemini choose 10 scenes that would be impressive for our announcement and add them to the post."
The oddest thing about this, to me at least, is that they pick that as their showcase. Did they themselves not notice? Or did they think their potential customers wouldn’t? Not sure really which of the two is worse.
That entire clip looks like it's straight out of a complete bargain-bin videogame. The weird smoothness of the ride, the pivot in mid air, the wings apparently don't have collision with the poles they fly past.
Definitely wouldn't have been one of my picks to show off my fancy new model.
Getting good quality dragon POV flight data for SFT must be really hard.
the goal posts have definitely moved since the original will smith spaghetti video huh?
Oh yeah, don't get me wrong "bargain-bin videogame" is a wild step up from "accidental amorphous horrors that defy description". It just jumped out to me as looking unusually bad compared to the other samples they provided. The stained glass deer, for example, was awesome.
The labs are the ones moving them. We're all just measuring against their own, exponentially growing, claims.
There's also a whole extra bonus castle that appears to the left when the camera looks left around 0:22. Which is amusing both because it wasn't there on the initial approach, and also because nobody puts two castles each with their own distinct curtain walls right next to each other like that.
"Storm the castle!"
"Which one!?"
I was under the impression he has already fly past the catapult.
It is always incredible to me how quickly we go from "this entire technology is pure science fiction" to "nitpicking over details"
It's not, it's just that hacker news is a hyperbolic bikeshedding chamber.
> Genie 4 has advanced audio capabilities
> Prompt asks for saxophone
Looks inside:
> Sound of a clarinet
Maybe Genie 5 will get it.
A few years from now, Lydia in Skyrim will have a fully consistent personality with dynamic dialogue and full contextual memories of all our adventures together in Skyrim.
A visual makeover is great but what happens when millions of people fall in love with virtual characters who seem to deserve that love?
skyrim already has mods for AI chat with all of your followers. I haven't tried it but plenty of people use it, (in VR as well, of course).
Question is if we even want that. Video games often are escapism, we want an illusion of reality but not actual reality. I mean there definitely is a market for that, but I don't think it's for everyone.
That actually sounds implementable even now! Except the fully consistent personality part. And maybe that's a good thing. Falling in love with beings without physical presence sounds agonizing; probably better spent on real people.
> Genie 4 is natively multimodal, enabling audio-like music, ambient noises, and interactive dialogue
What on Earth is audio-like music? Do they mean music-like audio?
Audio like: music, ambient-noises etc
That was probably meant to be an em-dash, or have spaces around the hyphen.
probably meant "audio like music, ambient noises, and interactive dialogue."
or just "music-like audio"?
Have you listened to "music" created by non-specialized models that use WebAudio to create?
It seems like a good descriptor to me! ;)
As opposed to video-like music...
I'm blessed to see that even google let their most important marketing prosa be generated by ai models without any proofreading from fleshy humans.
That isn't a mistake AI would make though -- it's something a human would accidentally do when replacing em dashes with hyphens
Which they would do if the text was LLM-generated and they wanted to disguise that fact. Or have a vibe-coded script to do the same.
It could have been a human trying to justify their existence. "no no, I wrote this ad copy myself: look, no em-dashes!".
What if we're all in Genie 10 right now?
Isn't that a simpler explanation of reality than the zillions of particles all existing simultaneously in this massive universe?
Some sort of simple primal generative AI construct that just feeds a conscious experience for one. You. Nothing else is real. No one else is real.
Something like that would be an infinitely simpler, than running the physics and conscious experiences of this unimaginably large universe we think we're in.
Yeah 2020 was the upload point when we were all corralled during covid. And trees were the data centers of the dinosaurs. I mean, what if?
Looks cool, but it's very hard to get me excited about yet another proprietary model for which absolutely zero technical details are released (not even the number of parameters), which is safetymaxxed to the limit, and whose usage quota will (most likely) be so limited that it'll be effectively useless for anyone who doesn't work at Google or doesn't have infinite money.
What an insane set of requirements for a frontier technology and your excitement by it.
i think he's bummed out that a cool thing has shipped but: 1. he doesn't get any information about how it works 2. he doesn't get to use it
sure, it's still a cool accomplishment but i can understand being a little bit frustrated / not as excited for such an announcement.
Most cutting edge technologies through history have been 1) somewhat opaque and 2) narrow/exclusive in their deployment. Historically speaking, that's virtually synonymous with "cutting edge technology".
Whining that I, a random Internet user with limited budget, usecases that the researchers don't care about, and no connection to the researchers, don't get technical details (??) or access (??) on day one is just profoundly goofy shit.
Obviously being able to use these things right away would be more exciting. But to be unable to get excited by advancements we can't immediately use is... quite entitled.
The sole purpose of a press release is to generate positive interest. It is not entitled for the reader to expect the time they have invested in reading it to be worthwhile.
Press release announcing a new car:
> erhmygord I can't even buy this car right this second for nearly nothing and use it to drive a thousand miles per hour through a kindergarten parking lot?!?!
If you mean that this press release amounts to the AI equivalent of the latest Tesla Roadster 'release'... then I guess you're kinda right
This.
And I wish I had thought to grab that nick years ago…
When tweetles battle beetles… AND…
Most cutting edge technologies through history did not have near-peer widely-deployed open competitors. This technology does.
It’s not an exciting advancement if it’s just the AI equivalent of google wave. It’s totally reasonable to be skeptical of marketing and want to try something for yourself.
Absolutely no need to slander Google Wave like that.
Oh, but there is, good sir. There is.
Fake hacker thinking Google Wave wasn't released to the public lol
Google wave was eventually released to the public but initially it required an invitation. So if you had a friend at google you could get in but if not you were out of luck. This killed early momentum and when they eventually opened up the product was dead on arrival.
> But to be unable to get excited by advancements we can't immediately use is... quite entitled.
Well, I didn't go knocking on Google's door asking them to give me access to their fancy models. It's them who came up to us saying "hey, look at the amazing model we have here; too bad we won't tell you anything about how it works nor even let you access it". Entitled you say? Then maybe don't dangle a carrot right before my face, and then complain that I would like to eat it?
Then don't click on the link homie
"Don't talk publicly about things I personally cannot use right now" is peak entitlement, yes.
> "Don't talk publicly about things I personally cannot use right now" is peak entitlement, yes.
I never said anything like this? My point wasn't that "I can't use it, therefore it's bad"; my point was that as-is there's pretty much nothing exciting about this. Just because this exists inside Google doesn't really mean or do anything; it doesn't meaningfully advance the state-of-art precisely because nothing about this is public.
Do you remember Sora[1]? Announced in pretty much exactly the same way -- grandiose webpage with fancy demos, zero technical details, limited availability and then... remind me what happened? Oh, right, they completely shut it down, and in the end it didn't "advance" squat. Might as well just never existed. So remind me why exactly I should be excited about this one here...?
[1] -- https://openai.com/index/sora/
In order to claim there is any 'advancement' here requires a baseline of openness which is not met here.
> is just profoundly goofy shit
Why not expect it? It's what most model releases do.
Yeah, I read it as: " Cool! Who cares? " Maybe because that's how I felt
You must have a very narrow conception of sanity.
The future is coming by appointment only.
I think this might be what Gibson meant by "... it's just not evenly distributed"
> frontier technology
It's just a lame gimmick, this is not 2024 anymore, people have started to get sick of this AI thingie, especially when it comes to video/image generation. That Simon guy with his pelican is the exception that can only be seen on tech forums like this one.
As to why US (and probably also some Chinese) companies have started pouring hundreds of billions into AI-image generation is beyond my comprehension, but on the other hand I'm no economic genius.
Not insane at all, completely sensical.
If I can’t realistically use it then I don’t care, especially when there’s no independent proof yet it works this well, given a cursory glance. If it trickles down into future developments I can use then cool, but it likely won’t because they want a moat. You don’t have to shield the poor defenseless multi-billion dollar megacorp from the genuine apathy they inspire.
If this is the best “frontier” they can achieve, then it’s no wonder they haven’t done anything groundbreaking in the space yet that produces practical results.
No it’s not insane.
It's another closed world model.
Only open weights matter.
We should distill everything else we can into open weights.
These are exciting for robotics and industrial perhaps, but they have no consumer use case until the lid is blown off and consumers can do whatever they want with them.
So much societal progress is behind held up because the leading labs want to keep a stranglehold on the tech. Imagine if every PhD or hacker had access to the raw weights.
The general history of new technologies is that they're available for a narrow set of circumstances and then they proliferate. Can't believe that needs to be said on "hackernews"
Default internet culture is "I am entitled to everything digital, in full, for free".
Which stems from the fact that most people have never created or contributed anything meaningful.
The internet has become a privacy and rent seeking hell hole.
It used to be a lot different.
Look at companies trying to build more things that pull value into their private cloud offerings.
Except that most technology was built in a way that didn't require all of the humanity knowledge and labor to be sequestered and then regurgitated back as slop that we need to pay as a subscription from what essentially is a rent seeking mafia.
The only one pretending this is something that needs to be said, for whatever reason, is you.
amen
I mean this doesn't seem very insane? Basic info about the model, the ability to actually use it. Google in particular is awful about getting access to their models from the UX side: I don't even bother trying them out unless it's via openrouter.
Yeah, I want that too, including the weights and training material. But I am still exited by what one could potentially do with it, by looking at the prompts and generated videos, even if it is cherry picked.
I’ve decide that the best (and maybe only) way for me to interact with google AI models, is on the search results
It’s where I get the most value, and my expectations are very low, no attachment if it goes away
Of course they are mining me for data anyway, but they were doing that before already (and I do have a Gmail account, although I’ve almost completely stopped using Chrome - btw, it’s crazy that you can’t add plugins/extensions to Chrome on Guest mode)
> "which is safetymaxxed to the limit,"
I feel refusal'd just *looking* at this page.
I don't get how Google is so horrible at product. Let me play with this. Otherwise it's DOA.
That touches on the core reason not to get excited about these Google releases: They release something cool and then don't iterate on it. Instead they go off building another cool demo instead.
Gah. Go iterate. Listen to customers, adapt, iterate some more etc etc. That's how products are made.
Some companies are so default-alive due to essentially building a money printer that risk tolerance has been tuned down severely.
So Google is the new Microsoft but with better software engineering quality. And two decades younger. And they're still slowing down.
(Maybe I should reallocate some of my GOOG stock into ANT after all.)
I'm not entirely convinced the new models aren't just a rollback to a clean training data state after the current model has been released for a while.
In the clip with a marble bust dropped on a watermelon, at this point https://imgur.com/a/IHLnkUs the bust should've toppled over, if the physics were properly modeled.
This is a reflection of a complaint I frequently make about so called "world models". It's more accurate to call them video game models. They don't simulate the world, they simulate video games. For that reason a common issue with them is that the physics are video game physics rather than real world physics.
I agree but I feel like the exaggerated physics were intentional and fitting with the video theme. If they had wanted it to topple, they wouldn't have posted the video.
It also failed to properly model the physics of knights riding wyverns into battle.
Which is fine for a video game. Not at all fine if you want to train robots with it. Trying to combine both use cases into a single model is doomed to failure IMO.
I would not be so opinionated, modern coding agents derive from chatbot technology. Very different problem domain. They share language but language is universal so that doesn't say much. Then they figured out that the pretraining is a good starting point for RLVR. Who's to say a more cinematic imagination can't be a good starting point for a world model? It doesn't need to be the whole solution.
If they don’t get ahead of this our robots are going to be sitting ducks against knights riding wyverns smh
How often do things in the real world not behave exactly as you expected them to?
I feeling I get from this is that this could be the mental model for a robot in the real world, continuously corrected by actual events.
We all have a world model in our heads of what we think is going to happen, we adjust quickly (or not) based on what does. This is pretty incredible stuff.
> How often do things in the real world not behave exactly as you expected them to?
They wouldn't be called "Laws of Physics" if obeying them was optional. If your center of gravity is outside your base, you _will_ topple over. That's just how gravity works.
Busts rarely hang in midair at a 45° angle.
As an aside, why fall into the role of being an apologist for google-hype and Big Tech?
They were demonstrating steering.
So what, the world model failed to model physics. The thing they said it did elsewhere. It's a valid critique
yes. I feel Google's video physics has always been bad. slow, aritifical, unnatural.
That's the thing about machine learning over time, and the big hole in the whole world model idea. Time and time again we make simulations and train models and we think that loop will work. It never does.
And why not? Because the first model that can actually learn to handle the simulation not only can do the simulation internally too, but it can do it better than out simulation.
It's where (ex-)"AI luminaries" go to die. It's a way to make glacial, slow progress in the correct direction while never actually getting anywhere.
So we get to a point and people keep making complicated simulations of one kind or another ... and then someone (some model) who's never used the simulation, and doesn't need it for model training gets by far the highest score on every benchmark and everyone agrees the simulation was kind of beside the point.
It depends if the goal is a more accurate simulation, or if the goal is a tool for creatives to build games or content.
If it's the latter, I don't see a 'hole'.
My point was that there is also the possibility of AI intelligence improving to the point that there isn't a difference anymore. That whatever beats the next set of benchmarks (which of course won't be the simulator or world model) is so smart it is both a better solution to the problem the world simulator poses and a better world simulator all in one and a better tool for creatives to build games or content.
This has already happened to other ML fields. For example, there's not much point anymore to ASR models (just feed audio into LLMs) and the space that TTS models occupy is fast disappearing. Oh and it's not just that Qwen 3.6 is a good ASR in addition to being good at coding mario. Doing things this way, not using an ASR model is fundamentally more expressive. A whole bunch of companies started up on ASR + TTS "towers" (lots of models, Pipecat style) as they're called, and they're dead companies walking. Their tech has been surpassed by just bog-standard generic LLMs in a way that they will never get the upper hand again.
A chain of models approach, a separate ASR, will never do this: "you said you agree but I can hear you're sarcastic, so I'll explain again". Also getting a conversational tower to switch languages when the user does that (or when it makes sense in a language course) is not quite impossible but very, very hard. And yet feeding audio directly into Qwen will almost magically solve these problems better than I will ever do it (lightly finetuned). It's fundamentally better in every way (except perhaps speed, of course. Although arguably such a tower needs an LLM too, it'll just have way less tokens)
This point has long been passed for image recognition. Feeding a photo or screenshot into a full LLM along with a request on what to do beats the crap out of any OCR, image recognition, ... model every time these days.
And now there's also hints of world models just being bypassed. Just giving a finetuned model the last 50 video frames (surprisingly, that fits in context, especially at reduced resolution), movement statistics, motor power, status, ... of a robot and asking what to do actually works (with heavy finetuning, but I think it'll start working without). Well ... AND then distilling that knowledge into a much smaller model because B200 racks are more than a bit to heavy for a logistics robot to carry around. But we'll get over that.
In the man walking through the dressing room scene, the camera starts behind the man. And you can see that the left half of his body is wearing a grey suit... but the right half is wearing a grey dress. :O
That stained glass landscape video is pretty cool. I wonder if it would be possible to implement that effect in a real-time game engine.
Q: Is there a easy way for average people to recognize these live super-deep fakes?
I know I could run the content through a fingerprint detector. However, I'm a professional software engineer...not a normal user.
What's the ethical impact of this technology??
The non-coherence of the world is a pretty solid tell. Any time the camera pans away and then back, the scene changes. The model doesn't remember what used to be there dozens of frames ago and so it invents something new.
Gotta love the kraken destroying the city in the background of the Hong Kong lovers motorcycle scene despite nothing in the prompt about monsters. Filmmakers are still safe.
It prompts "gigantic alien tentacles that destroy the world."
I suspect you're reading the base prompt, but the video also demonstrates adding further prompts to change what is happening as the scene progresses.
I thought that was pretty cool.
It was the complete lack of fear or urgency on the part of the characters for me.
It was their eyeballs not drying out or squinting due to the wind for me.
I guess they used Genie 4 to make a video about people using Genie 4 in different contexts. This may have some implications.
If the checker clip is meant to be used to develop a robot to play checkers, then the physics is not remotely good enough. When the robot takes the first piece, the piece looks like it's moving up from the board, then suddenly doesn't. The last move also isn't valid checkers, unless I'm missing something -- the robot offers its piece to be eaten and the human doesn't take it.
Can't wait for the hyper-realistic customer service bots and deepfake propaganda videos.
With Anthropic going public, I expect a lot of model releases from the competition signalling to the market. "We are still around, please leave some money for us too"
Google is already public, has a larger more normie customer base, many products to integrate AI in with established enterprise accounts in the millions... For some reason theyre not valued at 50:1 revenue... I dont think all that many people are going to be that excited for Anthropic as a business, Its an escape hatch, using the public as exit liquidity while the founders can still make a 100bil.
The public is the final "greater fool".
I don't know. With each new release of these models, they compare side-by-side how they are better than the previous ones, but at the same time I feel less impressed by each new release. Somehow they manage to be a seemingly big step forward but at the same time I feel I have seen so much AI that I just look at the common failure modes and they are all still there.
It's easy for them to show demos of abstract things because they don't necessarily have to follow real-world rules, but anything modeled after real world falls short of expectations.
guys they are building digital cages so they can rob you blind while you are in your nice digital world, sorry we will ignore this stuff into oblivion
The prompts build impressive clips.
I am more interested in how to make consistent "participants" and "environment" with multiple clips so they look like one longer clip.
Will that require same dev/coding harness and spec-kit-like engineering flow?
Like create a spec on how a movie will look like, feed contexts, create tasks so that multiple clips look congruent?
I would like to see how it breaks, don't spend all your focus on how nice it looks with specially curated shots. At least give a few examples of what happens when it breaks from running too long. They mention that it can fail under "Understanding current limitations" but don't show any examples. Especially since they mention the cases of safety and responsible use, it seems quite relevant to help prepare people by showing them some examples of how it can break and so you an learn to spot it.
Their posted videos show plenty of failures. Read the rest of the comment section.
I don't know if I'm missing the point, but Genie 4 almost looks worse than Genie 3 in terms of player control? It doesn't even seem to respond to inputs, it just looks like they took some AI generated videos and put the key widgets on top.
Astonishing to me that a company like Google would be willing to publish such a terrible looking product demo. Interactive, but with completely made-up inputs that are ignored?
I think you're exactly right. This was already a problem with Genie 3, but they've appear to have gone all in on video generation over controllability.
The original Genie was a great piece of work with a real sense of purpose - to enable agents to learn about the real world in the way Dreamer or Schmidhuber's paper sketched. But now most of the people that drove that mission have left Deepmind (Jack Parker-Holder, Tim Rocktaschel, Demis) and it's become this instead.
I wonder if that's because the target audience is people directing/producing (commercial) videos, rather than games.
another POV is, what even are these experiences? whom are they for? it would take Google 10 minutes to hire all the best game developers for Android, and cost them like $2m, which is peanuts, and give them fancy Google Research titles. but instead, they're trying to get 100 google PMs, who don't give their kids screentime and who don't play any games the objective of making a product for Bart Simpsons. like these people don't realize that being good parents disqualifies them from making immersive first person shooter, "narrative" experiences.
> what even are these experiences? whom are they for?
One possible use for these is to facilitate lab -> this -> real world for robot models. Basically better than lab / algorithmically designed environments and cheaper / easier to train on than real life. Something that "looks" like real life and can scale, to apply RL for robotics to.
The more parsimonious explanation is that 100 PMs are even less qualified to make surgical robots, but at least they can download a video game and play it for 5 minutes. There's a great scene in You Can See Everything where Billy Evans is watching bio presentations on 1.5x speed, which obviously, makes him look like a huge moron - "knowing enough" bio is very bearish.
Anyway, let's take your theory to its logical conclusion: the product development between real and simulated soft tissue behavior is much bigger than Exponential Time being a hypothesis and being proven. There isn't some effective acceleraltruist path here. Dario Amodei was a biotech dropout. The thing you are betting on which Google is not doing anyway, maybe they will destroy the descriptive single cell sequencing paper mill, but they're not going to be able to simulate the fat inside a BMI 30 abuela undergoing a gallbladder removal.
Sorry, I thought you were asking about the tech applications. I have no idea about the other stuff, can't comment.
Just as JPod by Douglas Coupland predicted.
When Google's epitaph is eventually written, PMs will figure prominently as the catalyst for the rapidly accelerating decline.
Why hire crew or actors when you could have them be fully AI? Why film a commercial with fluffed, oil dusted, well-lit hamburgers when you could fully render something? Why actually have a vehicle capable of off-roading when you could just AI-render what it looks like off-roading (or vertically taking off, for eg)?
It's a dystopian business use case that I hope will get obliterated in the near future as the generations that believe in this stuff and are fascinated by it age out of the population.
The point is simulation for autonomous vehicles and robots.
The future of movies will be promising.
[1]: https://www.minid.net/2026/8/2/the-actor-was-the-expensive-p...
all the things you've dreamt of - your vivid dreams - now you can replay them.
for all the a.i misery with hardware prices i.e ram.
such a wonderful world we live in now.
I highly suspect that Genie is being made primarily for internal use, and access is more to gather feedback/training data and a bit of PR.
I fully expect to see the same cat twice soon.
What is the point of this vs cheap robotic arm simulations?
The only place I see this applicable is for visual artists prototyping
Humans interact with their environment by subconsciously simulating it and predicting what will happen in the next few moments. This ability gives us the capacity to work more efficiently and reliably in any environment. These world models will serve a similar purpose to the AI and robotic systems of tomorrow. This is a foundational piece.
We don't have to fall off a cliff to learn to drive on a mountain road, we can imagine the consequences for cheaper. If we didn't have imagination life would be much more dangerous.
You could produce a thousand cheap ads for your product or one expensive real ad.
Podcasting is going through the thousand cheap ads with the obnoxiously inserted interstitial ads. What they have accomplished is me unsubscribing to the worst offending podcast and beginning the journey of ad blocking on podcast with the help of whisper and LLMs. Also now I hate Shopify with the heat of a 1000 burning suns due to their obnoxious ads and the insane rate they get played.
So yea that one expensive ad would probably have a positive impact compared to a tidal wave of slop. Quality > Quantity after all.
You could produce immersive VR content specified only by prompts and in which any action is possible.
Wish fulfilment gets extremely boring, extremely fast.
Video games are interesting because of very specific constraints the developers impose on the player.
The authors could still design their prompts to steer players towards predefined developments and situations. Anyway, the possibilities of something like this are infinite, the only constraint will be the cost of the compute.
>audio-like music
My favorite kind of music.
Again, very impressive from a technical point. But language is so flawed to do anything creative. It just turns out all looking the same and kind of boring
A nit, but given that this is from a major corporation, and LLMs are supposed to be em-dash-happy, it's surprising (disappointing?) that the announcement uses:
> audio-like music
That's a hyphen. What's "audio-like music" -- music that is almost sound but not quite? Music made from raw audio? :-)
Yeah, I think that sentence was supposed to be: Genie 4 is natively multimodal, enabling audio--like music, ambient noises, and interactive dialogue--to be simultaneously generated with the world.
That's a proper use of a hyphen though, an emdash would be wrong.
But it's not, unless you know what "audio-like music" is?
I don't think so, and in fact an em dash would be much less confusing.
audio, e.g. music
audio — like music
Oh maybe I did just misread the intent of the sentence? I read it as "something like audio" but yeah that is confusing since that would be all music.
That's a suffix word though, not a pause.
Hyphens should be used to link compound/suffix words, not em dashes.
Agreed that hyphens should be used to indicate compound words, but you're saying that "audio-like" is an intended compound word? Then it's an adjective, modifying "music"? What would that mean?
That was my initial reading and I may have just misread it as them intending it to be compound rather than an aside modifier.
Well okay, not a "compound" per se, but it's a prefix/suffix, which similarly should use hyphens.
I think the disagreement here is not whether that’s how you should use hyphens but whether that’s what was intended at all. The term audio-like music makes no sense.
I think you're fundamentally missing the point being made here.
Yes, "-like" is commonly used as a suffix, in which case a hyphen is correct. But describing music as "audio-like" is clearly nonsense (music isn't "like" audio, it is audio), so that's not the sense in which it's being used here.
For clarity, the full sentence is:
> Genie 4 is natively multimodal, enabling audio-like music, ambient noises, and interactive dialogue to be simultaneously generated with the world.
Why does everyone hate colons?
audio: music, ...
> Genie 4 is natively multimodal, enabling audio-like music, ambient noises, and interactive dialogue to be simultaneously generated with the world.
Looks like a hiccup, which created an odd and ambiguous sentence.
There used to be editors who would read press releases to catch this.
Amazing!
Question I have is when will they be able to render something more then 30 seconds.
And how much will it cost? There has to be a HUGE amount of computing power behind this. As a video game developer I live in fear and anticipation.
>coming soon to Google AI Ultra subscribers in Project Genie
Would be fun to muck around with world models. Are there any that I can try for under £200 p/m without being part of a special project?
Techincally, isn't the less expensive sub for Gemini Ultra £75 p/m?
It looks sooo boring and generic that it's hard for me to imagine anyone thought https://www.youtube.com/watch?v=sFfPkUJd_Jk is a good showcase.
The playable element is cool PR but the news is that endless synthetic training data for embodied AI models is basically here or close.
And their most impactful application will be training smaller models for military drones to detect, target and kill humans. sigh...
Every time I'm about to be excited about AI the cynic in me slaps me back to reality. I imagine the endless possibilities of AI for disabled people, then realize that it's going to leave millions of people disabled instead.
It will probably do both
Surprised that "do you really think we're going to make it?" scene made the cut of the first video, her expression and tone is totally out of sync with the scene and demonstrates the problem with almost every AI gen video I've come across when it comes to capturing humanity.
The painting->interactive scene looks astonishing though
Neuralink integration and I'm out of here.
Google aren't the only ones with this tech. You can try out a 24/7 rolling set of video feeds for yourself at https://fal.live
An LLM generates four alternative prompts for what could happen next and there's a real time vote running for each option. It's trippy, like watching someone's dream.
Showing off an anglo-bias with the Japanese train station in the 1980s, with a man speaking english.
Infinite Tsukuyomi is coming soon…
Good moving toward open world:)
It's hard for me to understand Google's redefinition of "world models" (basically: a txt2videogame modality for LLMs) as anything other than an attempt to bury Yann Lecun's definition of "world models", which includes an implicit bear case against LLMs and thus Google's stock price.
So world models capture inherent complexity of physical reality to help teach AI about hidden laws which probably helps train both LLMs and multimodal models used for autonomous vehicles and robots? And AMD acquiring World Labs is a move to outdo Nvidia’s work in this space. And Google probably sees this as a gateway to a new platform of autonomous vehicles and humanoids that they could rent out as a CSP? Competing with Nvidia, Tesla and complementing Waymo. Am I getting that right?
That all sounds right to me, but apparently they can also be used for simply generating videos, or "unlocking the creation of new interactive experiences"... which both seem like competing goals with everything you just mentioned.
> "A REAL, ultra-photorealistic, high-end 35mm cinematic first-person perspective..."
Real nice prompting there, Google. It also still looks like a video game rendering.
ironically, the word "realistic" appears next to game footage, not next to real one.
Not ironic at all. You describe a rendered image as photorealistic, you just call a photo a photo.
That's the irony. The over prompting ends up doing the opposite thing it is intended to do.
The third video is proof that v4 can, ekhm, walk slower?
If it’s not open weights, no one should care.
It doesn’t matter how impressive your demo is if it takes a literal data center to run.
Actual innovation can’t be seen at the let’s just throw money at it level. But, it can be seen at the that’s cool. I can run it on my laptop level.
Hi, I always loved watching Pixar movies, partly because I knew that what took them days to render (in their literal data center) would eventually be done in real-time on my desktop.
In the same way, seeing that something is possible gives hope to the people who are making those open models that you want.
Failure modes?
Show them or I'll assume this is pure marketing.
You know you can already use previous versions of Genie and they look pretty damn close to this right?
pretty cool !
So if those genie 4 characters have all individual chain of thought reasoning and persistent retrospection / recollection, how do these human characters know they are not living in a simulation made by biological humans? Not sure you understand what i'm getting at. Some ethical questions do arise.
Is anyone actually making any games with these tools? Like, can I play one on Steam or something?
Impressive slop. But slop nonetheless.
Why can't Google just announce something and release at the same day, not always this stupid wait list stuff.
wait list, then it releases, and you get 10 api requests *per day*.
Until you've spend $100-150 on their platform. Then you get 250 per day until you spend $2000 and it becomes properly unlimited.
It is really annoying. I'm literally trying to pay you for a service and you won't let me.
Prompt: make a press release that isn't a slideshow on this company's own browser and hardware platform.
"Building world models responsibly"
... I'm suddenly reminded of Iain M Banks' "Surface Detail".
Who'd be down to write a science fiction novel where creating fake images is a crime and not a product?
I guess we'll find out when they publish it.
Hmm, wonder if Databricks trademarked their Genie AI product.
So does this export 3d assets or an Unreal/godot/unity project so it can be turne dinto a game or is it limited to their runtime?
i am starting to get model fatigue..
What the fuck are the little wasd/arrow keys supposed to mean? They have no relationship to the movement of the character or the camera. It's just flashing buttons.
What is this for? I cannot understand these models. They are very much polluting the internet with nonsense and use capacity that people and businesses need for actual work.
If these are the cause of why I have to pay 2-4x I used to for a computer, then Google should go fuck themselves.
why does this need to exist?
this is the gta6 I am was hoping for
Not really impressive at all at this point. That dragon clip was pretty awful. If you told me this page was from 2024 I’d believe you.
From the auto-generated captions on the YouTube video on this page.
> Genie form [sic] can simulate scenarios
It's kind of amazing: the juxtaposition of such advanced generative AI and such terrible speech-to-text coming from the same parent company.
The volume of content on YouTube is enormous, I’m kind of surprised they offer auto captions at all given that they have little to no competition in the space.
Is it surprising? Indexing for search is the backbone of Google's business model.
It's why if you search a practical task you will often see results link to specific parts of a video.
Ok but like, surely the Deepmind team could sneeze and get flawless captions for this video. Why rely on the default YouTube captions instead of just attaching correct ones? It seems silly.
Yeah if they cared they could create their own captions for sure. They probably didn’t even think about it. Maybe some intern will do it after reading this thread
They're not proof-reading the blog posts either. The second sentence:
> Genie 4 is natively multimodal, enabling audio-like music, ambient noises, and interactive dialogue to be simultaneously generated with the world.
(Emphasis mine). Like, yeah, I'd sure hope the music is similar to audio.
I wonder if this was a de-LLMese pass gone wrong, and the original was "enabling audio -- like music, ambient noises, and interactive dialogue -- to be simultaneously generated with the world.
When you're high on vibes, the line between music and audio-like music tends to blur.
These days, unproofread text adds authenticity.
Only when it's a mistake that a human being would plausibly make.
Typos mean the text came from a human.
[dead]
As I said before, World models will become simulation engines of AAA games where there is no need to release the actual game artifact and even if the game exists as a binary artifact, it will never be released and will be available on their hosted cloud back-end, then simulated and streamed back to you.
So, Stadia + World Models = New generation of DRM for games.
Why? Why would a non-deterministic engine be better for a game when it can be instead structured and made deterministic and then more effectively distributed to consumer hardware?
It wouldn't be better for players. It would be better for companies because the user no longer owns the computer that runs the game.
You can still stream a deterministic world to the player.
And a month or so of development time. Pure IP re-selling.
October? Yeah, just release the "Spiderman saves Halloween" prompt.
This is a wet dream for big studios.
it's not, it just sounds cooler to say prophetic things like this these days
Game industry is still struggling to find a way to incorporate regular LLMs without suffering catastrophic edge cases. A long way to go before these can do more than fancy tech demo, I think.
None of this makes sense. Think of the economics involved here: is it cheaper to run tens/hundreds of thousands of GPU clusters to simulate a unique game world for millions of players, or would it be cheaper to just write the god damn code once, and then host it for download, or to allow them to be streamed?
Because it wasn’t bad enough to need a $5000 5090 to play a game, now you’re going to need a full rack of them!
They can try, my household won't be purchasing those.
We'll be riding the huge wave of indie titles that are incoming.
> We'll be riding the huge wave of indie titles that are incoming.
The problem will be separating the wheat (people with good stories and concepts enabled by Ai) from the chaff (people with no taste slopping out trash with Ai)
Yes, but for me the discovery of treasures is part of the fun.
I also love thrifting and artifact hunting.
Isn't that what streamers are for?
I mean - reviewers!
streamers and best-in-genre videos have become my primary filter tbh
been checking out the archives lately, nothing recent excites me
Because stadia was a real banger right?
Do you have even the slightest idea of the level of expectation that even the most casual gamer has?
This isn’t achieving much beyond prototyping. They’re burning so much money for no reason.
And? Just spin up some open world model and pay for the compute and play your own games?
There are so many options for games now that just sticking with the biggest AAA companies seems like a waste.
Stadia is dead, the reason is playing games over the internet sucks, input lag and serious graphics degradation - this is why I cancelled
Ai generated games suck even more, no skill; no taste, there is no TLC put into them
Then there is the economics of it, cost prohibitive to (re)generate a scene and ship that over internet pipes
Stadia is dead because if you have a data-center of GPUs you are using them for AI. It has nothing to do with input lag and graphics degradation.
Oh no my guy, you are misremembering, stadia died a quiet death long before AI.
The latency was no good even on fast internet for even slow games. It is and was pointless.
Stadia was pretty good, be it OnLive was similar quality in like 2010. It even had the best version of Cyberpunk 2077 when that launched. But the model was free version of the service for a limited resolution AND you have to buy games there OR a subscription based option and you still have to buy games there.
That made no sense and wasn't a viable business model.
stadia died way beforehand, history is useful
https://www.denofgeek.com/games/why-google-stadia-failed-shu...
I am sure the are other typical Google fumbles that would have meant they were not the victor; but I am sure an equivalent service would be mainstream today if AI had not dramatically changed the economics. [also Google bought DeepMind in 2014; 'way beforehand' is and odd way to look at it]
I disagree, dropping below 60 FPS will be noticed and turn players away, you are not getting that with good input handling even today, it has nothing to do with GPUs and everything to do with networking
> beforehand...
is about the craze, the ChatGPT moment and beyond, when GPUs demand went exponential, not when various types of Ai algorithms were available
It is about when Google realized GPUs were about to go exponential; they may have missed some opportunities but they were well ahead of the curve.
sure... despite all the other people who used it and are saying otherwise, and the fact that google totally fumbled in the Ai race as well as built custom silicon for Ai...
... history is helpful ... google didn't - and still doesn't - understand gamers, same for all the other Ai hucksters, gamers are among the most anti-Ai
https://www.youtube.com/watch?v=c1Fv1uKTd-w
Geforce Now is still a thing
[dead]
And nobody will play that trash, and the world won't change.
Also, lmao, they're already trying to nickel and dime you and sell you microtransactions that took 5 minutes to make for 10 bucks. If you think they're going to pay the costs for running a world model for every single person, lol.
Y tho?
Every increment of advance in this field is an increment of evidence that we are simulations.
No it doesn't, not even close. You are looking at a video represented as pixels changing on a flat screen. None of it actually exists. You exist, the computer or phone in front of you is composed of mass and energy, and absolutely none of it is algorithmic. World models, as complex as they are, are still just pushing numbers around on paper.
Even if we can show that it is technically possible to model a universe like ours to the degree that our current perception of reality suggests exists, there are three other big factors that would need to be weighed:
Is it feasible? What Dyson sphere level of power would be needed to maintain such a simulation? Is that level of energy available to whomever is running the simulation?
Is there a positive ROI? What does this advanced civilization get out of running such an expensive simulation?
Is it moral? The simulation involves sentient creatures who are capable of deeply contemplating their own existence, capable of feeling rapturous pleasure and indescribable pain and loss, and ostensibly capable of making moral choices that range from heroic to depraved. Doesn't there come a point where the beings in a simulation are so close to the beings in "base reality" that the same moral principles apply? If you press a button that kicks off a simulation where humans can torture and murder each other, behaving essentially identically to humans in "base reality," aren't you in some sense facilitating those deaths?
> What Dyson sphere level of power would be needed to maintain such a simulation
Why would you assume energy is relevant to whoever turned on our simulation? If I created a simulation where the only source of energy was cotton candy, my creations might ask "but where would our creator even get enough cotton candy to run a simulation like this", not realize that it's absurd to think that the creator of a simulation has any of the same constraints that exist inside the simulation.
You basically described the plot of World on a Wire.
You run it to see if your military tactics will defeat the enemy and then you keep running it until you find a strategy that guarantees victory. Who cares if some sim lives are churned through the simulation, our lives are at stake!
Reductionist but I feel like that’s how it would go.
How does that argument work? I've seen more than a few advances, yet I don't see how any of it provides evidence that we are simulations. This might be one of those things that comes down to the definition of "simulation" I guess.
Not sure if you're joking, but the only such evidence I could think of would be something like finding some way to influence the hardware of our simulation and seeing a glitch in our world
[flagged]
Who is asking for this? Do advertisers really want this? I feel dread about many of the AI products being sold. They're mostly tools to help funnel money from the users to the bosses.