201–210 of 718 posts
Earlier quoted context omitted.
Not if they cannot get the GPUs.
If China can't get the GPU's they'll build them. If they can't build them they'll smuggle them, and either way, eventually they will take over Taiwan, and then the US will be wondering what they can cook with leftovers.
Earlier quoted context omitted.
Mark Cuban in a recent interview answered your question: companies are afraid there is going to be just one in the end—sort of the way there is one ad-company now on the internet. They want to be that one. Whether they're correct that there can be only one is of course a matter of debate. But that is at least the mind-set they are operating under according to Cuban.
Why would Mark Cuban know anything about the motivations of today’s big tech companies? He has not been involved in tech businesses since he sold a radio on the internet website 26 years ago.
> where it matters most: commercialization. It begs the question because both its premise and assertion are already wrong. Has AI improved the industrial capacity of the US in order to improve the lives of its citizens? No it hasn't. Has AI increased the wealth of its citizens by being able to do laundry or any household task in a generalized way? No it hasn't. The only thing it's really done is make very narrow slic…
You better have some sources for declaring that industrial capacity hasn't increased. The Fed reports around 8% penetration of AI in manufacturing already, but in my opinion it's too early for grand declarations like that without data. https://www.federalreserve.gov/econres/notes/feds-notes/moni...
Based on a survey if the business uses AI "in any of its business functions". And for all uses of what they consider to be AI, not just LLMs.
Inclined to disagree. The winner here will be whoever can move atoms with AI not take notes at the daily standup. i.e. Think boston dynamics vs unitree They're both doing well but I'd lean towards China is winning on atoms in light of a huge manufacturing base they can AI-ify.
You can tell we're on the cusp when level 5 self driving cars are common an you have multiple companies deploying them on the street. Google is doing great work but the poured TONS of effort into it and the thing still needs intense stacks of perception and processing. Much more than I've seen any humanoids pour into it.
L5 SDV's are much easier to get than humanoids and the have tangible economic benefit. My thesis is that those will come first.
Earlier quoted context omitted.
That seems like a lot of rationalization to me. China is pursuing these because they cannot compete on the frontier. Yes, there is a possibility that all that compute is not needed, but it is a rather remote possibility, and there is no doubt that, given the choice, China would be pursuing frontier model building with closed, propietary-only offerings.
China is competing in value AI because they cannot work at the frontier, but how is this bad at all? It’s like how the USA has the best drones but they are a few million dollars apiece while China has DJI. If China could work at the frontier, I don’t know, I kind of think they would still be dumping a lot of resources into exploring the value side since they have that culture already in place.
Article title: “The US is winning the AI Race” Article content: “The US are capitalizing on AI the best” A lot of assumptions there that no one can actually verify as true right now. If commercialization into rent-seeking SaaS landscapes is the endgame, then yeah, the US is winning the AI race. If individualization, local LLMs, and consumer hardware are the endgame, China is winning the AI race. If it’s something ent…
No, the US is _leading_ the AI race, but the race isn't over. What's the point of leading the race for 90% of it, if they're gonna slip on their own sweat and fall down by the end? In non metaphorical terms, what's the point of spending billions of dollars rushing to get the best AI tech at all costs, when the competition can distil your progress and catch up in 6-12 months while only spending 1% of what you spent. E…
What if it is not winner take all? What if there is no race. What if what USA has been doing is just burning money with possibly unsustainable debt load and way over build valuations... AI being commodity server capacity might be a thing. And the customers might even manage without hyperscalers... In that sort of end scenario whole current market might look rather foolish.
You mean, what if the hype-based billionaire-class is wrong? Isn't suggesting that a sin in America these days?
Earlier quoted context omitted.
The non-release of Mythos tell you the future of that, so long as they can keep the weights from being exfiltrated. Once models become true national security threats, they won't be released in their full form. The hitch-a-ride approach becomes less capable of keeping up.
How would they prevent distillation? That would seem pretty tough to block for any LLM available for commercial use.
Honestly, I think its quite possible that models will be retrained with gaps in their knowledge. e.g. a coding model for commercial use probably doesn't need to have deep knowledge of biology, and training on biological sciences probably doesn't help those evals much.
Honestly, I'd welcome such an approach.
It's like people forget the entire point, perhaps even definition of technology is "doing more with less."
The "brute force" of power and cycles is almost certainly the least important thing, perhaps even a hinderance.
Earlier quoted context omitted.
This. I dont know what the benchmarks are supposed to represent, but to me Kimi K2.6 is indistinguishable from e.g Opus 4.6.
How much would one have to invest in hardware to feasibly run a usefully large Kimi K2.6?