-
0:04:37 · watch on youtube.com↗
Their wordsIt turns out that the proper computer, the perfect computer, is one that could do sequential processing and parallel processing, not just one or the other.
- 1 min later
-
0:05:58 · watch on youtube.com↗
Their wordshaving it being a large market is important because the technology is complicated. And if we had a large market, our R&D budget could be large.
- 1 min later
-
0:06:30 · watch on youtube.com↗
Their wordsA GPU is like a time machine because it lets you see the future sooner.
- 6 min later
-
0:12:05 · watch on youtube.com↗
Their wordsWhen you create something new like CUDA, if you build it, they might not come. And that's always the cynic's perspective. However, the optimist's perspective would say, but if you don't build it, they can't come.
- 5 min later
-
0:17:33 · watch on youtube.com↗
Their wordsyou reason from first principles. And at some point you have to believe something. And if those principles don't change and the assumptions don't change, then there's no reason to change your core beliefs.
- 2 min later
-
0:19:09 · watch on youtube.com↗
Their wordsBut you have to believe in your future. You have to invest in yourself. And we believed this so deeply that we invested tens of billions of dollars before it really happened.
- 1 min later
-
0:20:32 · watch on youtube.com↗
Their wordsThe fact that model size and the data size being larger and larger, you can learn more knowledge, is also true, empirically true.
-
0:20:59 · watch on youtube.com↗
Their wordsThen the question, the only other question is, what can you learn from data? What can you learn from experience? Data is basically digital versions of human experience.
-
0:21:27 · watch on youtube.com↗
Their wordsSo we've now demonstrated that AI or deep learning has the ability to learn almost any modality of data, and it can translate to any modality of data.
- 2 min later
-
0:23:11 · watch on youtube.com↗
Their wordsthe last 10 years was really about the science of AI. The next 10 years, we're going to have plenty of science of AI, but the next 10 years is going to be the application science of AI
- 4 min later
-
0:27:37 · watch on youtube.com↗
Their wordsIn order for robotics to be smart about the physical world, it has to understand things like gravity, friction, inertia, geometric and spatial awareness.
- 3 min later
-
0:30:21 · watch on youtube.com↗
Their wordseverything that moves will be robotic someday, and it will be soon. You know, the idea that we'll be pushing around a lawnmower is already kind of silly.
- 1 min later
-
0:31:45 · watch on youtube.com↗
Their wordsAnd so I think the idea that we'll have our own R2-D2 for our entire life, and it grows up with us, that's a certainty now.
-
0:32:11 · watch on youtube.com↗
Their wordsWe should talk about what could go wrong so we can keep it from going wrong.
- 2 min later
-
0:34:34 · watch on youtube.com↗
Their wordsAnd so the AI safety systems has to be architected as a community such that these AIs, one, work, function properly. When they don't function properly, they don't put people in harm's way, and that there's sufficiently safety and security systems all around them.
- 1 min later
-
0:35:47 · watch on youtube.com↗
Their wordseverything in the end is about how much work you can get done within the limitations of the energy that you have.
- 1 min later
-
0:36:19 · watch on youtube.com↗
Their wordsWe're far from having any fundamental limits that keep us from advancing. In the meantime, we seek to build better and more energy-efficient computers.
- 1 min later
-
0:37:47 · watch on youtube.com↗
Their wordsAnd that's essential because we want to create more intelligent systems and we want to use more computation to be smarter. And so energy efficiency to do the work is our number one priority.
- 2 min later
-
0:39:36 · watch on youtube.com↗
Their wordseither one, that transformer is the last AI algorithm, AI architecture that any researcher will ever discover again, or that Transformers is a stepping stone towards evolutions of Transformers that are barely recognizable as a Transformer years from now. And we believe the latter.
- 1 min later
-
0:40:18 · watch on youtube.com↗
Their wordsThat's in fact the essential beauty of a computer, that it's able to do something today that no one even imagined possible 10 years ago.
- 2 min later
-
0:42:15 · watch on youtube.com↗
Their wordshaving a computer that enables the flexibility of research and innovation and and new ideas is fundamentally the most important thing.
- 1 min later
-
0:43:02 · watch on youtube.com↗
Their wordsEven though we have them made by TSMC, we assume that we need to have the deep expertise that TSMC has.
- 4 min later
-
0:46:35 · watch on youtube.com↗
Their wordsAnd if we could see the future and we can predict the future, then we have a better chance of making that future the best version of it.
- 4 min later
-
0:50:07 · watch on youtube.com↗
Their wordsAnd so I think that the next decade, intelligence, not for everything, but for some things, would basically become superhuman.
- 1 min later
-
0:50:37 · watch on youtube.com↗
Their wordsAnd I'm surrounded by thousands of them. And yet, what is it? It never one day caused me to think all of a sudden I'm no longer necessary. It actually empowers me and gives me the confidence to go tackle more and more ambitious things.
- 1 min later
-
0:51:42 · watch on youtube.com↗
Their wordsAnd if there's one thing that I would encourage everybody to do is to go get yourself an AI tutor right away.
- 1 min later
-
0:52:12 · watch on youtube.com↗
Their wordsWe're going to become superhumans, not because we have super— we're going to become superhumans because we have super AIs.
- 4 min later
-
0:55:59 · watch on youtube.com↗
Their wordsAnd learning how to interact with AI is not unlike being someone who is really good at asking questions.
- 2 min later
-
0:57:40 · watch on youtube.com↗
Their wordsThe next generation doesn't have to ask that question, but it has to ask the obviously next question, how can I use AI to do my job better? That is start and finish, I think, for everybody.
- 1 min later
-
0:58:10 · watch on youtube.com↗
Their wordsIt is easier to prompt ChatGPT to ask it anything you like than to go do the research yourself. And so we've lowered the barrier of understanding.