If you're depending on the money generated by UA-cam you have to follow that pattern. I hope he maintains the quality of his videos. No shortage of quantity of ai news
To elaborate: those of us in the know, understand GPUs are suboptimal for inference. The scenario is similar to how GPUs were initially used for crypto mining, before more specialized ASICs came out.
So which of these two scenarios faced more risk: 1) A Jensen Huang's $40K investment in his startup Nvidia in 1993? 2) Or someone today buying Nvidia stock on the public stock market?
Blackwell isn't more power efficient its just two previous gen chip fused together. The efficiency comes mostly from lower floating point accuracy. Groq is built from the ground up to be inference efficient. Blackwell is great for training terrible for inference and that is where the rollout to end users will be will accelerators (NPU). Nvidia lead is mainly in-situ graphic cards and drivers running CUDA libraries. This will not help the mobile market.
From a different perspective, the U.S. is the best place for A.I. investment and research. It's energy and capital rich, with large domestic sources of oil and natural gas free of geopolitical risks, unlike Europe and China.
No, but there isn’t much there and the U.S. also has the potential to easily tap sources of geothermal energy in certain areas, but we’d much rather protect them while oil and natural gas are easily accessible. States like Texas are also making progress on wind and solar, and microwave energy is being developed to allow for deep drilling to access sources of geothermal energy without the heavy investment of replacing drill bits every few meters. So, the U.S. has the sun, the capital, and lots of fertile farmland to feed lots of skilled people to develop industries like A.I. What country were you talking about again?
@@BrianMosleyUK Ahh, there we go… I can always expect non-Americans to hit below the belt when they can’t come up with anything else. I will vote for whoever prioritizes the interests of Americans, whether it’s Biden with his Buy American policies or Trump with his America First policies. People like you are the reason why I didn’t buy chorizo because it was made in Spain and bought one made in the U.S. instead.
Speaking of energy and capital, where was the world's biggest solar farm just completed? Giving 5GW of energy, enough to power AI clusters orders of magnitude greater than currently available?
![The Meteoric Rise of Nvidia [Fastest Growing Stock]](http://i.ytimg.com/vi/nVtb2vNUOdU/mqdefault.jpg)
You're such a great communicator Pete. Massively underrated channel. 🙏👍
The Neuron is by far my favorite source of news for AI. Everyone else is just milking the passion everyone has by making unnecessarily long videos
they really upped the kawaii on this channel and i love it haha
Agreed. There are like 3 good ones that I can think of and this guy is one of htem
One day we tell people we subscribe to the neuron before 5k and they won’t believe us
If you're depending on the money generated by UA-cam you have to follow that pattern.
I hope he maintains the quality of his videos. No shortage of quantity of ai news
Now this is how you turn complexity into simplicity! You have a new subscriber! Thank you!
Pete! Pete! Pete! The man is back 😎🙌🏻. This was super interesting 🔥
Great content! Keep up the good work! Subscribed!
Hey man, I thought your intro and graphics looked pretty damn good!
The new graphics are so fresh! Really impressed by Clipt
Hey pete,
Why not make a video about available ar glasses
GPUs may not necessarily be the substrate for AGI/ASI.
To elaborate: those of us in the know, understand GPUs are suboptimal for inference. The scenario is similar to how GPUs were initially used for crypto mining, before more specialized ASICs came out.
So which of these two scenarios faced more risk:
1) A Jensen Huang's $40K investment in his startup Nvidia in 1993?
2) Or someone today buying Nvidia stock on the public stock market?
Blackwell isn't more power efficient its just two previous gen chip fused together. The efficiency comes mostly from lower floating point accuracy. Groq is built from the ground up to be inference efficient. Blackwell is great for training terrible for inference and that is where the rollout to end users will be will accelerators (NPU). Nvidia lead is mainly in-situ graphic cards and drivers running CUDA libraries. This will not help the mobile market.
🤡🤡🤡🤡🤡
Global Domination
The cat is cute !!!
❤❤
LOL the people that voted on this were all investors ;)
From a different perspective, the U.S. is the best place for A.I. investment and research. It's energy and capital rich, with large domestic sources of oil and natural gas free of geopolitical risks, unlike Europe and China.
Have you ever been to Iceland?
No, but there isn’t much there and the U.S. also has the potential to easily tap sources of geothermal energy in certain areas, but we’d much rather protect them while oil and natural gas are easily accessible. States like Texas are also making progress on wind and solar, and microwave energy is being developed to allow for deep drilling to access sources of geothermal energy without the heavy investment of replacing drill bits every few meters.
So, the U.S. has the sun, the capital, and lots of fertile farmland to feed lots of skilled people to develop industries like A.I. What country were you talking about again?
@@o_o825 don't forget you have the most stable democracy ahem... Good luck following your next election.
@@BrianMosleyUK Ahh, there we go… I can always expect non-Americans to hit below the belt when they can’t come up with anything else.
I will vote for whoever prioritizes the interests of Americans, whether it’s Biden with his Buy American policies or Trump with his America First policies. People like you are the reason why I didn’t buy chorizo because it was made in Spain and bought one made in the U.S. instead.
Speaking of energy and capital, where was the world's biggest solar farm just completed? Giving 5GW of energy, enough to power AI clusters orders of magnitude greater than currently available?
Cuda
This is a side show. AI is fastest developing in China, Russia and maybe Japan.