Just like being in a dream. Trying to run towards "that hill over there" only for it to get further away as you take each step, or it disappear when you take your eyes off it for a second. It's astonishing to have that very private experience (that I've never really put into words or shared with another person) be rendered in front of me by a neural net.
Watching this era of AI advancement play out is like stepping through a museum of the inner workings of your mind.
Interesting. I read somewhere a long time ago a description of dreams being incoherent because 'the brain doesn't have enough compute power to simulate reality well enough'. This seems to me like the exact experience you're sharing. Reality is generated, but no 'memory' to keep it consistent/coherent.
So, are dreams really just GenBrain (and the things you think about before falling asleep the generating prompt)?
I would guess that this is correct. My dreams are very much like half way between waking consciousness and a surreal movie - my consciousness doesn't seem to play a part in generating the surroundings and scenery and most of the plot, that seems to be the result of GenBrain.
I think daytime thoughts are also GenBrain btw. Only that waking brain has access to higher mental function - namely longer term memory, higher propensity to notice/surface inconsistencies etc.
Funny, I just used it on Firefox for Android and it worked.
Impressive, but practically unplayable because, for example, the moment grass comes near to the camera it will likely halluci-spawn a wall or something right before the camera (behind the grass). Every time I saw a place to walk to, I was never able to reach it because of similar effects. But: nice idea.
In this case, the reason might be that it needs WebGPU, which is only available in chromium based browsers. Changing my user agent doesn't let me in either. Text selection being disabled might be a "lazy" fix for text being selected while in game.
It's fascinating how walking backward will completely make it forget the previous locations and sometimes even change biomes entirely. Anyway it's still insane that we can do all of this in real time.
"Diffusion Models Are Real-Time Game Engines" was a paper from August 2024. How am I playing a real time AI generated game in my browser in 2024? I did try uploading a random image of a random flow chart I had, and it ended up being mostly noise... but still, how?
Readit News
Watching this era of AI advancement play out is like stepping through a museum of the inner workings of your mind.
I think daytime thoughts are also GenBrain btw. Only that waking brain has access to higher mental function - namely longer term memory, higher propensity to notice/surface inconsistencies etc.
> Currently, our app is available on Desktop Chrome only. Please switch to Chrome for the best experience.
The best experience? I guess seeing this message technically is an experience.
Shame they didn't make a website for people to use though. Oh well, next tab!
Impressive, but practically unplayable because, for example, the moment grass comes near to the camera it will likely halluci-spawn a wall or something right before the camera (behind the grass). Every time I saw a place to walk to, I was never able to reach it because of similar effects. But: nice idea.
The future will be wild. "Hey ChatGPT, lets play Counterstrike on the Enterprise-D. Counter-Terrorists agains Spongebobs"