What happens if you lay off 80% of your department while your competitors don't? If AI multiplies each developer's capabilities, there's a good chance you'll be outcompeted sooner or later.
At some point soon, humans will be a liability, slowing AI down, introducing mistakes and inefficiences. Any company that insists on inserting humans into the loop will be outcompeted by those who just let the AI go.
> "Get bankrolled by the state at the state's discretion until they get what they want, even if they need to burn $1B to get $1M of value"
If that's how it worked, they wouldn't lead in anything, they'd be bankrupt already. They burn state money like VCs burn cash. DeepSeek, Alibaba, Tencent, Xiaomi, Huawei, etc., disprove your point.
Look into how their 5 year plans have lead to capital investment with almost zero feedback. A heavily bureaucratic system of bureaucrats incentivized to spend massively to boost their own appearance, and cover up losses/inefficiencies.
Ghost cities, empty high speed rail lines, solar cells being mass produced at a loss.
All these things also produced end products the state wanted, no doubt. But the capital allocation strategy is basically a "throw all the money the leader gives in that direction until the leader says stop".
Is there a lot of wasted capital? Sure but a lot of it still produces outcomes.
> A heavily bureaucratic system of bureaucrats incentivized to spend massively to boost their own appearance, and cover up losses/inefficiencies.
In China, if you want to move up politically, you generally need to show results, meaning the province or area you govern is expected to deliver measurable performance (even if politics and connections still matter too). In that sense, you could argue it's more performance driven in some respects than the US.
EVs and solar were clear priorities, and China has been very successful at scaling both and driving costs down. Domestic competition has been so intense (especially in EVs) that margins have gotten extremely thin, and officials have recently signaled they want to curb "irrational" price wars.
> Ghost cities
Sure, some exist, but many of the developments that were circulated online years ago have filled in over time. That said, there's no question a lot of projects stalled or collapsed during the property downturn, especially after China Evergrande and other developers ran into trouble.
> empty high speed rail lines,
I can't speak to every route, but overall the high speed rail network is heavily used. When I traveled in China, it was excellent and extremely extensive. Some lines and stations likely see weaker demand than others, but the idea that it's broadly "empty" doesn't match reality.
> solar cells being mass produced at a loss
With overcapacity and price wars, many firms have faced serious margin pressure and losses though that doesn't mean every producer is losing money on every panel.
In the end, the real question is whether the capital allocation is efficient enough for citizens to benefit and for the country to remain competitive. Empirically, the answer looks closer to yes in industry and infrastructure, while real estate has been a major exception, with real costs and inefficiencies.
Not really. China doesn't share a border with us, doesn't claim any EU territory, and didn't historically rule our lands the way the USSR did. In the context of spheres of influence and security interests, its strategic goals aren't directly at odds with the EU's core interests.
> EU is not a singular country, and Germany or France don't border Russia either.
But soon they could, that's the problem.
> Considering China is ok to supply Russia, I don't see how your second point has any standing either.
Supply? China supplies Ukraine too. Ukraine's drone sector runs heavily on Chinese supply chains. And if China really wanted to supply Russia, the war would likely be over by now, Russia would have taken all of Ukraine.
Years ago I got interested in D. It's a great language, but at the time its garbage collector was leaky. There weren't any D entries on the Benchmarks Game back then, so I ported most of the programs to D and optimized them as best I could as a newcomer. Performance wise, D was in the C/Rust/C++ range and in many cases it even beat Rust and C++. I tried to get the community involved to help the language gain wider adoption, but nothing really happened. I think everything has its moment, and D's moment has passed. They didn't make the most of the window when D could have gone mainstream.
Honestly, this is the most reasonable comment here, especially coming from someone in Taiwan. I hear similar views when I'm in Asia, which are very different from what I hear back in the West.
> Plenty of researchers do think so and claiming consensus for your position is just false
Can you name a few? Demis Hassabis (Deepmind CEO) in his recent interview claims that LLMs will not get us to AGI, Ilya Sutskever also says there is something fundamental missing, same with LeCunn obviously etc.
Europe is one of the world's largest agricultural producers and exporters. France alone is one of the top grain exporters globally. The EU exports massive quantities of wheat, barley, dairy, and processed food to North Africa, the Middle East, and Sub-Saharan Africa. Countries like Egypt, Algeria, and Nigeria are heavily dependent on European grain imports. An AMOC collapse would devastate growing seasons, slash yields, and potentially make large parts of Northern Europe unsuitable for current agriculture.
And it's not just food. Europe is a major producer and exporter of fertilizers. If European industrial and agricultural output collapses, the ripple effects hit global food supply chains hard. Countries that depend on those imports will face famine.
Then there's the knock-on, hundreds of millions of people in food-insecure regions losing a key supply source, simultaneous disruption to Atlantic weather patterns affecting rainfall in West Africa and the Amazon, potential shifts in monsoon systems affecting South and East Asia. It's a cascading global food security crisis.
> lots of time to adjust
This assumes a gradual slowdown, but paleoclimate evidence suggests AMOC transitions can happen within a decade or even less. The idea that we'd just smoothly adapt to one of the most dramatic climate shifts in human civilization is not supported by what we know about how these systems behave.
Because gathering training data and doing post-training takes time. I agree with OP that this is the obvious next step given context length limitations. Humans work the same way in organizations, you have different people specializing in different things because everyone has a limited "context length".
They couldn't do it because they weren't fine-tuned for multi-agent workflows, which basically means they were constrained by their context window.
How many agents did they use with previous Opus? 3?
You've chosen an argument that works against you, because they actually could do that if they were trained to.
Give them the same post-training (recipes/steering) and the same datasets, and voila, they'll be capable of the same thing. What do you think is happening there? Did Anthropic inject magic ponies?
reply