The man was sad. Then, he saw that a two-minute paper had been uploaded. Upon seeing this, the man became happy. - this text was generated by ChatGPT for mature audiences.

depressing to hear

get a life

Bisquik

ANNNND AND AAAND AAANNNDD

Microstrategy CEO bought $155million worth of bitcoin, so yes BUY!

It’s going to be a wild year for these sectors, so you should def. invest in crypto. 60% of my portfolio is spread across tech stocks, crypto and Crypto/Gold ETFs.

well the crypto market is expected to do way better than any other equity sectors this 2024 especially with the SEC crypto ETF approval but it’s a volatile market nevertheless and if you’re new to it, it’s best to reach out to an experienced adviser for proper guidance.

Yes, my asset manager advised I spread further into mutual funds and crypto Etf and boy am I glad I did. The whole idea is: Don’t get too greedy and also to exit at the right time, so generally I do find having an adviser very helpful, because what Avg. Joe really has time to watch and comprehensively analyse the market.

could you recommend some good advisers? don’t get me wrong, I already have an asset manager, but he seems not to know much about crypto.

Felt a bit uncanny valley to me 😜

yeah and the design fits it's leather jacket

Very human design

meh

Terrible times to be alive. Nobody a part from billionaires will benefit from these technologies. Millions of job will be lost, millions of people will be useless.

silence and keep working 9-5

​@@PuppetMasterdaath144 nah, these things are created to do your job 24/7.

​​@@PuppetMasterdaath144WHAT A TIME TO BE ALIVE!!!

Quality not quantity as they say. ;)

same for AI Explained.

7B: 130B models, “AI UKpostsrs”: ??

@@vectoralphaAIAI Explained is usually one of the first channels to release worthwhile content, however his content is a little bit more on the technical side.

The visuals are information-dense, but not the audio. The images are compilations of demos. The audio is half catch-phrases and almost no real explanation. I watch these videos because I want to keep up with what AI can do--but I'm under no misimpression that I understand _how_ AI is doing it.

Buy stocks related to that sector 😊. Eg battery, wiring, steel etc related to these AI giants.

I feel like both the costs and the wattage requirements to output ai will decrease over time. This reminds me of the airline industry were it was a race to make the biggest plane and now its about improving the efficiency

We need better Fusion generators

not just electricity but it burns through water too

it's fine, AI fill figure out a way to solve the problem

I actually get a little depressed if a new mind-blowing AI breakthrough doesn't get announced each day.

​@@TheManinBlack9054It really might be. I'm starting to think when people look back there going to put the start of the singularity back in the mid 2010s or farther back. Techs been like this for awhile even before the pandemic it's just totally out of control now

I don't know about that. Everything is relative and we tend to adjust to things rapidly. The advances in AI, however incredible, aren't actually rising as rapidly as the amount of papers suggest. The models haven't changed drastically in recent years, but been fine-tuned. The underlying systems are more or less the same, just that they perform better and we've found ways to streamline the processing. In terms of science, there haven't been much of huge real scientific revalations that would change our perception of science or the world around us and counterintuitively, we have a massive problem in the scientific community. The amount of papers have risen exponentially, but less and less papers are following scientific principal and most of the papers are literally trash. And I'm talking about natural sciences at this point. There's surprisingly little we can find out, unless we find out something that changes our understanding of nature entirely, which seems very unlikely since Einstein have been on trial for ages and there's millions and millions of ways relativity have been tried to be proven wrong.

@@eliteextremophile8895 I see. Thanks for the long comment! I can see this being a problem in fields like quantum mechanics and astronomy. My field of chemistry does still seem to be having some pretty amazing discoveries being made all the time. I think AI will be a major part in making the discovery that changes our understanding of the world entirely.

@@noob19087 ah, good point and thanks for your reply. I did overlook the field of chemistry in my comment. It seems that spontaneous stochasticity spawning from molecular level interactions reaching wide spread, even global effects is true and as long as we can't observe most of molecular level noise, it seems very likely that we won't be able to predict highly chaotic systems very well, which will undeniably have long reaching limitations to what chemistry will be able to provide, but that was pretty much the span of my thought in this context. I truly haven't had much thought about the possibilities of chemistry due to my tendency to physics, but now that you mentioned that field, I feel like I really should. There might very well be a way to find chemical interactions that can be highly coded to behave very intricately to mimic natural stochastic behavior, but more predictably, resulting in natural coding of sorts. Thanks for bringing me the chemistry perspective in this context, I will have a deeper look into it. What do you think? What new perspective chemistry could bring us in our exploration of natural world?

I think 20 years is beyond our imagination..IT IS fast🤯

@@thegoodsherman 1 year is beyond imagination, it's fascinating, I prefer to be optimistic tho

Question is how amazing you think it's today compared to years ago. I just see a lot of tools that speed up things but not that easy to master and we get easily bored of and want the next thing. A lot of uncanny valley. We have more freedom than ever to built something online but still a lot people are worried about the future and guessing wildely about it.

My prediction is that in 20 years we will see complete real time environments all made up with AI (text to image diffusers type of thing but in real time), it would be able to simulate stuff you tell it to, it would be able to work as a game engine, it would be able to be experienced in VR/MR in realtime with 6 DOF working with it so we would be able to move our head or rotate it in the VR/MR realms and experience whatever we want with it, AI would figure out stuff on its own and would invent stuff on its own and would solve any given problem, maybe there would be a main chat bot that would have access to many many more specialized AI bots that would be able to work together to do stuff the user asked for in the original prompt, we would also have simulated realities in video games in VR/MR and on screen that would be like actual living beings that would live their lives and would think stuff and so video games would become way more realistic with their NPCs and way more interesting as result, each user would get a different storyline and would be able to create and develop storylines with these NPCs like in the real world, also robots would become way more realistic and smart and capable than they are today, using advanced type of Isaac Gym kind of virtual training of robots, also the AI would design robots from text for example so it would be - text to robot and then it would be ran in Isaac Gym to make it work in the real world (after it would also be created according to the AI blueprints and instructions, maybe there would be a robot builder which would be a special machine connected to this AI model), also by then in 20 years I think that there would be also AI model that is able to generate other AI's so text to AI AI model so you would describe, advanced AI like AGI or probably even ASI by then would able to also discover new materials and sciences ETC, maybe the AI would even invent new types of cultures it would come up with that it would develop over time.

@@MrRandomPlays_1987 sounds possible😵‍💫🤯🕺

Watch Tesla's Optimus. That's not just a toy, but a fully functional humanoid robot, ready for mass production. It's still being trained, but even that shouldn't take long now.

By that definition we've been in a singularity for the past 500 years. Beware the hype machine. Notice that no part of any of this was about making people's lives better. It was all about efficiency. They're selling this tech to corporations, not you.

@@michaelleue7594 Never said it was making my life better, just that it was amazing to see. In general, technological singularities are not viewed as good things for humanity, quite the contrary in many depictions. For the AIs, though, maybe it's a different story.

This is not even the beginning of the singularity. That's a whole different ballpark.

@@Rick104547 The transformation have already started when people care less about exploring real life and are stuck in social media and clickbait news adrenaline. It said to be perfect then it must be, it doesn't matter if it exists yet, the showcase says it's going to be wonderful. No pain no gain, if everyone can do anything it has little value, but it requires more to go your own unexplored way. There are lot of creative AI tools that take time to master.

@@Rick104547 I shouldn't really have to justify my statement but I will attempt it anyway. I have watched this field (AI) my whole life and it was always such a completely intractable problem, up until very recently the attempts were almost laughable. Then, suddenly, with LLMs the problem was cracked, and progress has been blazingly fast since then. We have seen how LLMs can now write computer code at a professional level, it is only a matter of time before they start improving themselves, something that will likely happen in the next few years. From there, it's out of our hands and we get Singularity. Barring any attempts to rein in this technology I see nothing to prevent this from happening. Time will tell.

These are all existing things, Nvidia is just providing a platform for them. Their main product is still the hardware, which is ultimately just a commodity. The software platform is a great help for small dev teams, but the big boys use their own anyway.

Now if they could just make some reasonably priced graphics cards

All of these are old news. Nvidia only in the hardware side and making a very good job of keeping the hardware out of reach for 99% of the people. So, I deeply doubt what good GTC is apart from increasing Nvidia's share price.

@@aniksamiurrahman6365 They don't sell to you, they sell to Microsoft, Google, Meta, Tesla, and other large corporations with practically infinite money. They have like 70% margin on the AI training chips. And their new platform is also for AI training.

@@nic.h Price is what people are willing to pay.

5:20

You will be fine. We don't know much yet, it could be just another marketing strategy by nvidia.

More likely a new AI-designed "pandemic" within two years.

The entire world should ensure not to purchase a single 6GB or 8GB graphics card from NVIDIA's 5000 series, and that'll teach 'em to stop giving us anything less than 12GBs going forward. Give us 24GBs. Give us 48GBs.

I want a couple TBs of VRAM within 5 years @@imjody

​@imjody Nvidia isn't a charity, they don't "give" you anything. 8gb is more than enough for almost any game at 1080p.

@@Wobbothe3rd it's 2024. Try gaming at 1440P or 4K.... 1080P? Gross.

​@@imjody go outside

this has been known for thousands of years

The probability is 99,999...%.

Simulation of what

Not currently but we will in a few years

@@Instant_Nerf yes.

Now, get ready to be laid off.

@@aniksamiurrahman6365Yes I'm already thinking about that.

That does impose soft boundaries on the timescale and the detail of the predictions, but it doesn't imply that conventional weather-prediction doesn't have substantial room to improve.

@@mvmlego1212 But it kinda does. At this very moment our weather predictions are very good and the reason sometimes predictions are good for weeks but sometimes not even for hours is due to this exact problem in butterfly effect, not because our methods aren't good enough. Sure there's room to improve, but I doubt it's substantial.

You just gotta feed it more data.

prolly got past joiner in the masons and decided they love screwing folks up the poo hole for eternity like everyone visible, Ever

and it's reaching its peak if we do not find a way to make computer truly creative in the true sense of the word. Which it can't until we find a way to build a computer that is truly creating new stuff based on spontaneous stochasticity. We need a computer that isn't based on gate arrays. That's pretty hard cause that's the only way we, at the moment anyway, think computers have to work. And to be honest, I much rather wouldn't want to create a god. Which AI would be if it had true creativity of human and calculation power of computers.

@@eliteextremophile8895 human creativity is the same as the NN Models, we train our brain with our 5 senses and then try to recreate things that makes us feel that again. isn't that what art is all about?

Just submitted half the video (20 instances) of that as "filler" via the SponsorBlock extension. It's just a shame it's so time-consuming to do that for every video...

​@@Gerlaffynone of them are seen on my end

@@HokoraYinphine none of the skips? You will need to make sure you have the "filler" skipped in settings

I was at GTC. I’m pretty sure I saw someone controlling the Wall-E robot

Oh, well if ​@@dna2dna275 is “pretty sure” …

they can be manually controlled but don't need to be

@@dna2dna275 Thanks for clarifying, I guess they were partly AI like the movement which is fine from my perspective.

@@Jamgcy What does fully AI even mean.. we get tired of systems faster than ever and want to build the next version that have more features.