China does not have a way to deal with crime in Cambodia. It does not have anti-terrorism law to operate in other countries, also it does not want to upset Cambodia or Myanmar government when not necessary. These Chinese operates in Cambodia are mostly on the wanted list anyway, they don't plan to go back to China. In fact, a person leaving China to Cambodia and Myanmar will be checked and make sure their trip is innocent.
Personally I hate these scammers, they have ruined so many people's lives. It seems that it will never go away. Too much money involved. I wish we can launch drone attacks on these places.
this theory would contradict with the national security reason used by the government though. I have read in Chinese sources that they plan to remove the European directors, reasons I don't know.
It seems that microchips are a major export, a single industry in a country collapsing can cause a lot of damage. From other mentioned articles, it seems like the company was setup to extract all value to prop up another company... not dissimilar from K-mart and Sears stripped of all value at the end. Now imagine that accounts for say even 5% of GDP and other industries rely on that product as well.
I've been advocating for years that US prescription medications should require dual sourcing and at least 50% domestic production purely for security reasons. You can ramp up from 50% with multiple providers a lot easier than 0.
I can see that. WingTech bought it for profit, the best way to get the maximum reward is from the stock market. It is actually not rare operation in Chinese companies. But I doubt this is the reason Dutch government would interfere using the national security excuse. There are other ways to handle this, and would be less damaging. I am more inclined to believe that the US's pressure and the current management who wants to use this as a chance to take control of the company.
Is there a AI market for open weights? Companies like Alibaba, Tencent, Meta or Microsoft makes a lot sense. They can build on open weights, and not losing values, potentially beneficial for share prices. The only winner is application and cloud providers, I don't see how they can make money from the weights itself to be honest.
I don't know if there is a market for it, but I know that open weights puts pressure on the closed-models companies into releasing their weights and losing their privileged situations.
The only money to be made is in compute, not open weights themselves. What point is a market when a commons like huggingface or modelscope? Alibaba made modelscope to compete with HF, and that's a commons not a market either, if that tells you anything.
By analogy, you can legally charge for copies of your custom Linux distribution, but what's the point when all the others are free?
It promotes an open research environment where external researchers have the opportunity to learn, improve and build. And it keeps the big companies in check, they can't become monopolies or duopolies and increase API prices (as is usually the playbook) if you can get the same quality responses from a smaller provider on OpenRouter
we like to simplify things for the ease of understanding. I wish everything can be attributed to one factor, the world would be so simple and easy to understand.
That said, I do have my own understanding on what has contributed to China's rise, and I agree, the engineer aspect has a big play in it: I read the Lean Startup book, while I was reading it, I realize that a lot happened in the country share similarity to startups. Chinese government is very flexible, they do A/B testing with economic policies, pivoting all the time, the measurement, it is so interesting to see how concepts are very similar in both governing and running a startup.
By research you mean web search/deep research or only use knowledge embedded in llm? I use ChatGPT most of time, didn't find Claude work better for me, maybe I should switch if there is a big gap in performance.
Also not all compute was necessary for the final model, a large chunk of it is trial and error research. In theory, for $1B you spent training the latest model, a competitor will be able to do it after 6 months with $100M.
Not only are the actual models rapidly devaluing, the hardware is too. Spend $1B on GPUs and next year there's a much better model out that's massively devalued your existing datacenter. These companies are building mountains of quicksand that they have to constantly pour more cash on else they be reduced to having no advantage rapidly.
Ignoring energy costs(!), I'm interested in the following. Say every server generation from nvda is 25% "better at training", by whatever metric (1). Could you not theoretically wire together 1.25 + delta more of the previous generation to get the same compute? The delta accounts for latency/bandwidth from interconnects. I'm guessing delta is fairly large given my impression of how important HBM and networking are.
I don't know the efficiency gains per generation, but let's just say to get the same compute with this 1.25+delta system requires 2x energy. My impression is that while energy is a substantial cost, the total cost for a training run is still dominated by the actual hardware+infrastructure.
It seems like there must be some break even point where you could use older generation servers and come out ahead. Probably everyone has this figured out and consequently the resale value of previous gen chips is quite high?
What's the lifespan at full load of these servers? I think I read coreweave deprecates them (somewhat controversially) over 4 years.
Assuming the chips last long enough, even if they're not usable for LLM training/serving inference, can't they be reused for scientific loads? I'm not exactly old, but back in my PhD days people were building our own little GPU clusters for MD simulations. I don't think long MD simulations are the best use of compute these days, but there's many similar problems like weather modeling, high dimensional optimization problems, materials/radiation studies, and generic simulations like FEA or simply large systems of ODEs.
Are these big clusters being turned into hand-me-downs for other scientific/engineering problems like above, or do they simply burn them out? What's a realistic expected lifespan for a B200? Or maybe it's as simple as they immediately turn their last gen servers over to serve inference?
Lot of questions, but my main question is just how much the hardware is devalued once it becomes previous gen. Any guidance/references appreciated!
Also, anyone still in the academic computing world, do people like de shaw still exist trying to run massive MD simulations or similar? Do the big national computing centers use the latest greatest big Nvidia AI servers or something a little more modest? Or maybe even they're still just massive CPU servers?
While I have anyone who might know, whatever happened to that fad from 10+ years ago saying a lot of compute/algorithms would be shifting toward more memory-heavy models(2). Seems like it kind of happened in AI at least.
(1) Yes I know it's complicated, especially with memory stuff.
(2) I wanna say it was ibm Almaden championing the idea.
I'm not the one building out datacenters but I believe the power consumption is the reason for the devaluation. It's the same reasons we saw bitcoin miners throw all their ASICs in the bin every 6 months. At some point it becomes cheaper to buy new hardware than to keep running the old inefficient chips, when the power savings of new chips exceed the purchase price of the new hardware.
These AI data centers are chewing up unimaginable amounts of power, so if nvidia releases a new chip that does the same work in half the power consumption. That whole datacenter of GPUs is massively devalued.
The whole AI industry is looking like there won't be a first movers advantage, and if anything there will be a late mover advantage when you can buy the better chips and skip burning money on the old generations.
The rule of thumb i heard was that over the "useful lifetime" of a chip the cost of energy+maintenance/infrastructure is about the same as the cost of the chip. IIRC the energy was some like 15-20% of the overall cost over the "useful lifetime". I'm putting it in quotes because it's kind of a cyclical definition.
That makes me wonder if it's more a performance thing than an energy cost thing. I guess you also have to factor the finite supply of reliable energy hookups for these things, so if you're constrained on total kwH consumption your only way to more TOPs is upgrade. Probably ties in with real estate/permitting difficulty too. I guess what I'm picturing is if energy availability(not cost) and real estate availability/permitting timelines weren't issues, that %20 of cost probably wouldn't look too bad. So it's probably those factors combined. Market pricing dynamics are hard :/
I didn't know the recycle time of the Asics was that fast! That's an interesting point about the first-mover. I would counter that a large part of the first move value in this case is ai engineer experience and grabbing exception talent early. But with Facebook and these guys paying for them like athletes, I'd guess that experience build up and talent retention aren't as robust. Openai lost most of it's best, but maybe that's an exceptional example..
All of that to say, yeah, first mover advantage seems dulled in this situation.
Who knows, maybe apple is playing the long game and betting on just that.
I recently rode a scooter for couple of times, and I find it is the best thing to move around in neighbourhood. It is convinent, parking is easy, cheap to run. Everyone should have one electronic scooter.
reply