Video

To be honest. The left hand side of clickbait picture is more likely showing a complete novice with no knowledge of how to use the settings already in Blender, and the right side shows the same picture after his 8 year old daughter told him about samples.

The voice in this video sounds like its voiced by an algorithm

Normally I automatically block any channel that uses the phrase "this changes everything", but as it's you I'll let it go.

Sword Art Online: Alicization predict this accurately!

WHAT A TIME TO BE ALIVE

fuck we're inside one of those games rn arent we

Ok now let's do this but with Rimworld

I've been playing with these things (ChatGPT etc) since early 2023, exploring their roleplaying capacities and I'm not surprised at all. We're creating what are essentially text-based "holo-decks". I've had some very uncanny moments when interacting with NPC. The hurdle is moderation because as the user one is essentially a god and you can alter anything.

Diagnostic AI assistants trained on pharmacology and physiology data with access to a clinic's patient visits, doctor's notes, and client's EHR could provide physicians with way better QA control in some cases, like during a day when a doctor's mind, body, or emotions might otherwise impair their decision making. For professionals stranded in 100hr/wk work conditions with increasingly smaller teams of aging out nurses, or even assistants in other industries/occupations, this could be a scalable, low-cost SaaS tool and improve quality of care across the board. Wait for VC to jump all over AI during the next low-rate bull run. We'll be seeing improvements everywhere in the next 2 years.

We are all AI agents in a simulation

this is so useless

If you dont have a relationship with the God that made the universe- You're 👏Going👏To👏Plug👏In. How crazy would it be if there was a book written thousands of years ago that predicted this exact thing?? But because you dont know Him, you'll willingly fulfill prophecy 😢

How do u play the game

Are they called TransAgents? 😅

4:35 You underestimate the genius of remote-bamboo-controlled rowing boats.

Can we see this improvement in the next nvidia drivers? Updated ray and path tracing for current games ?

Resumé: Ray Tracing STILL SUCKS

Would be nice if we could get an electrician, a carpenter and a plumber... LOL The Chat GPT towns run on magic... lol

ok and what is the point of this?

Ian Cheng did it first lol

I am more and more bored by AI.

I love your videos but because of the way you speak, I can’t help but imagine Ren, from Ren and Stimpy, teaching me all this cool stuff!

not people, but companies will be out of job...

yeah they literally steal their training data it's highly unethical

So chatgpt upgraded from being a single toaster to being a network of toasters. Unimpressed. So called "AI" is plateauing hard. Makes sense, since they are just glorified interpolators of objective functions in some space of such large arbitrary degrees of freedom that literally no one on earth has any idea how to control, more less say with any confidence how reasonable or accurate their predictions are.

I remember when the first RTX cards came out and Raytracing was just super bad and not even possible properly on the lower end cards. Like unless you had the 2080 or even better the 2080 Ti forget about Raytracing unless maybe you wanted to compromise on other quality setting or play in 1080p. And lets be honest you dont buy a 500€ card of the latest generation to play in 1080p, not even back then. DLSS didnt work well either. But we have come so far since. I'm not sure better graphics will really make the experience much better from this point onwards tbh. but if they are able to push things further thats nice. Especially if it works well with existing hardware and not only with their latest 5090 Ti for $2800...

Please make a tutorial how to create this myself

Had Gemini imagine is a TCG game developer. Through the exercise I had it create the most effective play-testing methods and also map out the interactions our devs would need to know. Reviewing it, it was more in depth that I would have guessed possible. Saved me a ton of time. All from playing the game of imagine you are a game designer.

Assuming people optimise their lives around tax policy is not particularly accurate...

What if life is all a simulation and we are all agents of the same AI?

How can I build this what will I need to build it

where the f is the simulation site link

Do you have enough power computation to run 25 chatgpts and your best idea is to organize a valentine's party? Really n*?

Looks like Sumotori Dreams Game 😂

Just need dozen 5000x improvement to make it into production flow.

❤❤❤❤❤❤❤ WATTBA

Meanwhile in a Virtual AI hospital Receptionist = Fatoumata Diawara.

I think it’s funny that you think the same chatbot can occupy different roles. As if humans don’t do the exact same thing throughout their own lives. Even siblings live very different lives.

this is how skynet begins. the ai is training itself on itself. it no longer needs a human to tell it what it can and can't do

Way ahead of them @mythmakerai

We're so fucked

Rip graphics designers

Hang the DJ episode of black mirror, where they can establish compatibility through repeated simulations. You have an AI persona in your profile.

I would watch a season of this simulation. I would pay to be able.to.create some complex sims and have them create their own drama or have significant conversations

I enjoy this because it really lends itself towards simulation theory... because it implies that many less-intelligent models working together (such as GPT 3.5) can be more effectively intelligent than a 'smarter' model (such as GPT4). Which means a simulated world with lots of 'dumb' people interacting is probably the most effective way to make AI do some work, not making them superintelligent, just cramming more of them together - IE, it actually makes sense to simulate a world like ours and fully explains pretty much all aspects of why it is the way it is I also like the implication that businesses are structured specifically in a way that they can get work done even when some or all of the employees are actually terrible at their jobs But, some previous research has indicated that having the AI second-guess itself, or use itself for a feedback loop, actually harms its performance significantly. It's interesting that this improves it at all

QnQ Everyone be using this voice now

I thought it was Cypher from Valorant speaking

We say this shouldn't replace real doctors. However, in some parts of the word this healthcare would be better than no health care.

The sad part about this research: the real goal is to optimize for insurance claims liability, not for healthcare outcomes. Example: while the described research in this video proves you get better healthcare outcomes if you get an early diagnosis and follow a particular treatment plan, the “other research” will demonstrate that you get less insurance claims liability if you deny the screening test and let the patient die early. Yes it is f”d up beyond belief, but such is capitalism. Unless we establish a robust auditing regulation for AI models, this is not going to end well.

Is this channel sponsored by NVIDIA?