WarChud

599 posts

WarChud

WarChud

@SheerC12972

TACO True Believer

TACO Bell Katılım Ekim 2025
254 Takip Edilen21 Takipçiler
WarChud
WarChud@SheerC12972·
@ChinaPatriot99 @YanLiuSZ @Bluebearmonkey @hyperphonk Yup, he should look into what "high quality growth" means in all the government reports. Its the opposite pumping real estate. In the new economy, science / engineering people win and finance/RE types lose. That's by design
English
0
0
1
16
словянский гиперфонк
"China is socialist" is such a cope. It's a hypercapitalist state obsessed with money, money, business, business, capital, stocks, investors. Social welfare is minimum. Workers rights are abused. VAT, IIT and Corporate tax are on average lower than in developed European countries
English
83
32
414
125K
Victor T.N.
Victor T.N.@VictorTN_83·
@GlennLuk @robert_baiguan It may not need quantum. They can simply leapfrog silicon chips and EUV machines. The Chinese are researching multiple approaches, for example, bismuth oxyselenide transistors and non-binary AI chips. If that happens, ASML and TSMC would become obsolete all at once.
English
1
0
6
499
Robert @Baiguan
Robert @Baiguan@robert_baiguan·
China will almost certainly do an "EV" on semiconductor. When that happens, the shock to the existing silicon economy would be a magnitude greater than the previous shock to the global auto industry. But even if this were to happen, it wouldn't affect today's market sentiment at all. People can't see that far. People can't imagine what they can't imagine. Unknown unknowns can't stop people from YOLOing all the way
Jukan@jukan05

One of Korea’s leading semiconductor scholars, Professor Seokjun Kwon of Sungkyunkwan University’s Department of Chemical Engineering, said in an interview with Korean media today that China could secure advanced semiconductor manufacturing equipment approaching extreme ultraviolet, or EUV, lithography tools by around the mid-2030s, despite being blocked from accessing them by U.S. export controls. He said, “We should neither overestimate nor underestimate China’s technological capabilities. We need to assess them coldly and objectively.” Professor Kwon especially warned that Korea should not take the long-term potential of China’s semiconductor industry lightly. He said: “During periods of industrial transformation, technologies that once seemed unlikely to be used can suddenly emerge. That is what disruptive innovation is. China’s electric vehicles are a representative example. China chose EVs as a way to overcome the long-established ‘moat’ built by the U.S. and Japan in internal combustion engine vehicles, and as a result, it has developed world-class technological capabilities. There is no reason the same thing cannot happen in semiconductors. Across China, Huawei fabs and industry-academia cooperation centers are simultaneously developing EUV alternative light sources, optical systems, and PR, or photoresist, materials. If one of these technologies survives, China could quickly move onto a growth curve backed by its enormous domestic market. A crisis could emerge in which Chinese semiconductor equipment, materials, and technologies begin to have a global impact.” Professor Kwon was particularly concerned about the possibility of China developing next-generation lithography technology. “Today, everyone says that cutting-edge processes below 5nm are impossible without EUV. But precisely because of that, EUV, monopolized by the Dutch company ASML, is also an environment highly susceptible to disruptive innovation. China’s accelerator-based light source technology could become the next-generation technology after EUV. A prototype could emerge as early as the mid-2030s. If that happens, even if SMIC remains around ten years behind TSMC and Samsung Electronics, it could eventually enter the single-digit nanometer process regime, meaning advanced ultra-fine process technology.”

English
7
25
238
32.1K
WarChud
WarChud@SheerC12972·
@0xdoug But it's funded by the cash flow and retained earnings of the most profitable companies the planet! They can definitely maintain these cash flows in the upcoming oil shock crisis when every basic physical good rises in price
English
0
0
1
278
WarChud
WarChud@SheerC12972·
@ai_hyperbull @The_AI_Investor @grok Go to OpenRouter, all top open models have many American hosted endpoints backed by VC funded companies targeting enterprise Also chinese models being good enough also means Google, Meta, xAI will catch up to anthropic in a year
English
1
0
0
22
AI Abundance Investor
AI Abundance Investor@ai_hyperbull·
For consumers, sure. But enterprise, especially US Fortune 500, will only work with American AI. They don't trust Chinese not to take their data or use their tokens to steal ideas. Anthropic would make even more money if they didn't have those Devs at $200/mnth compared to the enterprise payers in the $1m's
English
1
0
0
45
The AI Investor
The AI Investor@The_AI_Investor·
If Anthropic reaches around $300B in revenue by 2030, with inference gross margins of about 70%, it could generate roughly $210B in inference gross profit. If training compute costs are around $100B and ex-compute operating expenses are about $20B, Anthropic could produce roughly $90B in operating profit. This company could be worth a few trillion dollars in future.
The AI Investor tweet media
English
30
40
305
30.9K
Fold Universe
Fold Universe@folduniverse·
@zephyr_z9 Netflix, Hollywood, Bollywood and Korean wave gonna be disrupted by mIcRoDrAmAs? lol Nezha isn’t even a soft power and cultural export success as it’s just Chinese watching Chinese content. Even a small production like K-pop Demon Hunter can whoop their ass and win Oscars rofl.
English
1
0
0
642
包叔
包叔@baoshu88·
豆包收费,时机把握的很好,清理VPN撤梯行动,正值大部分国人无法使用ChatGPT、Claude,给豆包打开了收费空间
包叔 tweet media
中文
170
10
201
83.4K
Zac
Zac@Zac_Pundi·
@SheerC12972 Yes, Dola is also Bytedance owned. And Tikotok shop also bought Tokopedia. Can you imagine that?
English
1
1
1
1.8K
WarChud
WarChud@SheerC12972·
@abskoop 敢要这么高,小道消息很可能是真的 下个豆包模型会非常厉害
WarChud tweet media
中文
0
0
5
5.9K
ahhhhfs
ahhhhfs@abskoop·
豆包也要付费了 专业版 908 一个月🤡 一年 9488块🤡🤡
ahhhhfs tweet media
中文
665
21
566
551.1K
WarChud
WarChud@SheerC12972·
@pandawatch88 Or its that Mimo 2.5 pro posted surprisingly strong results in benchmarking over the weekend.
English
0
0
2
121
goldenlabubuwatch
goldenlabubuwatch@pandawatch88·
Most loved stock by shortsellers squeezed 8% up today from 52w low on either management updated GM% guide for 1Q26 (less bad than expected, given mix towards more expensive phones), or 50% jump in deliveries MoM, you pick. 3 weeks to results. None of the headwinds are gone: -tax breaks and subsidies for cars on the way down dont help -same with phones and consumer electronics subsidies flat/down -memory prices still up -internal r&d up big In the column of positives, just: -new models coming in (YU7 GT 1000bhp looks sick) but in a tough domestic market. International market, which has been BYD and Geely saving grace lately, is not a thing for Xiaomi until 2H27 In the column of wild cards -entering US with lifestyle products would be interesting -xiaomi 18 with embedded agent. Their AI lead is big believer in openclaw. Lets see what she cooks. Ah consumer electronics, never easy.
goldenlabubuwatch tweet media
English
3
2
31
3.7K
WarChud
WarChud@SheerC12972·
@JaredF449913 @morris_que14 Anyone can pump GDP numbers with mega stimulus. Keep felating those numbers. Keep pumping money to retirees and health workers to "consooom", GDP will look fantastic. When chinese government annouments mention "high quality growth" you should look into what it means. Bye.
English
0
0
0
12
Jared F
Jared F@JaredF449913·
If your response is insults, you’re not really engaging the argument. No one said “everything” China does is malinvestment. The point is marginal efficiency of capital is declining, which is measurable. When it takes 6–9 dollars of new credit to generate 1 dollar of GDP vs 2–3 before, that’s not a talking point, that’s a structural trend. Same with local government debt and overcapacity in sectors like real estate and heavy industry. China is clearly moving up the tech ladder. That doesn’t contradict the fact that parts of its growth model are becoming less efficient. Both things can be true at the same time.
English
1
0
0
16
WarChud
WarChud@SheerC12972·
@JaredF449913 @morris_que14 Okay you can keep fellating Chang and Pettis talking points. You're right, everything China invests in is malinvestment. Accounting identities are more powerful than moving up the tech ladder. No point debating further
English
1
0
0
20
Jared F
Jared F@JaredF449913·
Calling it “cope” doesn’t address the actual issue. Overcapacity and misallocated capital in China aren’t old talking points, they’re current and well documented, especially in real estate and local government debt. China didn’t collapse like some predicted, but that doesn’t mean the distortions don’t exist. You can have growth and structural imbalances at the same time. And this kind of debt-fueled, investment-heavy model isn’t new, it’s similar to the paths Japan went through before its slowdown and the Soviet system’s reliance on state-directed capital. And the efficiency trend matters. It used to take roughly $2–$3 of new credit to generate $1 of GDP. Now it’s often cited closer to $6–$9+. That’s a huge drop in capital efficiency and exactly what you’d expect from overbuilding and debt-driven growth. The debate isn’t whether it exists, it’s how sustainable that model is long term.
English
1
0
0
14
WarChud
WarChud@SheerC12972·
@fleetingbits My theory is that most compute is being used for research which diffuses as employees hop around Many OpenAI employees have moved to Chinese labs like the head of Tencent Hunyuan
English
1
0
2
75
FleetingBits
FleetingBits@fleetingbits·
@SheerC12972 it becomes a question of how are chinese labs outperforming an increasing compute gap or are we wrong about the existence of a compute gap
English
2
0
3
224
FleetingBits
FleetingBits@fleetingbits·
some quick thoughts on the us / china ai capability gap 1) caisi evaluations indicate that deepseek v4’s capabilities lag behind the us frontier by about 8 months 2) and, that the gap between us models and chinese models is widening over time, deepseek-r1-05-28 was only 5.3 months behind the frontier 3) now, i think we should always assume that ai capabilities are a function of available compute 4) so, when we see a graph of showing a capability gap, we should always look to see whether the available compute to each party supports the gap 5) and, when we look at this, we see that total us compute has been growing at a faster rate than total chinese compute 6) epoch ai has estimated compute for us hyperscalers and for china; us compute has been growing at ~5x per year; chinese compute has been growing at ~4x per year 7) in q1 2023, china was 10 months behind the united states in gpu compute; but by q4 2025, china had fallen to 20 months behind the united states in gpu compute 8) and, if we think that capabilities are log-scaling with compute, then we should expect the gap between the united states and china to be increasing over time 9) this is the importance of export controls by the way, we should expect that with more available compute, chinese models would be closer to the us frontier 10) and, if anything, the relevant question we should ask ourselves is how china is outperforming it's compute, since it's models are 12 months ahead of their available compute 11) if we see this gap narrowing, without the compute growth rate narrowing, we should ask ourselves what is wrong in our assumptions of capability growth
FleetingBits tweet media
Séb Krier@sebkrier

DeepSeek V4’s capability lags behind leading U.S. models by about 8 months. nist.gov/news-events/ne…

English
4
6
48
9.5K
WarChud
WarChud@SheerC12972·
@ai_hyperbull @The_AI_Investor @grok Open weight models released in April got really good at agentic coding. You're assuming monopoly level pricing power on something that will be heavily competed.
English
1
0
0
49
AI Abundance Investor
AI Abundance Investor@ai_hyperbull·
@The_AI_Investor This chart is already dated, they are up to $44B in May. They'll be at $100B by December or sooner. I see them going to $450B in sales by 2030 with $10T market cap. @Grok what would that P/E and multiple be based on the original poster's formula?
English
2
0
2
499
House
House@Gen__House·
@robert_baiguan Graph is perfectly fine, you don’t know what you’re talking about
English
2
0
0
600
WarChud
WarChud@SheerC12972·
@JaredF449913 @morris_que14 Lol this is cope that's been pushed for a long time. Gordon Chang was shilling this 25 years ago and nothing happened
English
1
0
0
14
Jared F
Jared F@JaredF449913·
Overcapacity is building a brand new bridge, and then destroying it and building a new one because you have government stimulus that came from a local provincial government that needs to hit 4.5% GDP target. Overcapacity is building ghost cities with 60-80 million EMPTY apartments. Overcapacity is building electric cars that no one buys, so you park 10s of thousands of them on vacant lots to slowly rot because no one is buying them.
English
1
0
0
17
Caius_Dean
Caius_Dean@Caius_Dean·
外头的ai都在“工业革命” 国产的模型还处于普及对话聊天 订阅制不被接受 没钱继续加大研发 只能玩降智的版本 还能“弯道超车”吗?
中文
2
0
4
341