Here to discuss is of the Silverado Policy Accelerator.

We get into:

• Why this is, in Dmitri’s words, “a disaster”

• There are military balance of power implications for selling chips to China

• Why the rest of the AI ecosystem is against selling chips to China, Why Trump made this call anyway, and why SME export liberalization might be next

• Where the GAIN Act goes from here

Listen now on YouTube or your favorite podcast app.

Jordan Schneider: Let’s first toast the unfortunate U.S. Attorney for the Southern District of Texas, Nicholas Jon Ganjei. On Monday morning, he proudly issued a press release for his cool-sounding “Operation Gatekeeper,” which intercepted $160 million worth of Nvidia H100s and H200s.

That afternoon, President Donald Trump announced on Truth Social that the United States would allow Nvidia to ship its H200 products to approved customers in China. Dmitri, please make sense of this for me.

Dmitri Alperovitch: There’s no way to sugarcoat this — it’s a disaster. This isn’t only about the Department of Justice. The U.S. Attorney General’s statement highlighted how critical AI is to military applications. The President’s own AI action plan discussed how the United States must aggressively adopt AI within its armed forces to maintain its global military preeminence, while ensuring that the use of AI is secure and reliable. This technology is essential to U.S. military dominance and the successes of the U.S. Intelligence community.

You have to give the administration credit — it is doing a lot to ensure all levels of the U.S. government are adopting AI. Why we would enable China to do the same is beyond me. Are we going to sell them aircraft carriers or Virginia-class submarines? Should we let them into AUKUS? This is effectively what we are doing.

It is outrageous that Jensen Huang has been able to pull the wool over the eyes of people in government and on Capitol Hill, convincing them that arming our primary adversary — the one we are unquestionably in a cold war with — is somehow good for America. I understand it’s good for Nvidia’s sales and for him personally, but it is a disaster for our national security.

Jordan Schneider: What I find baffling is the contradiction in Nvidia’s public messaging. Jensen Huang and his company argue that their technology will revolutionize every conceivable industry, all requiring massive amounts of GPU capacity. But when asked directly about the military implications of selling these chips, Huang downplays the risk. He suggests that China’s military will acquire the necessary chips regardless and claims they are too sophisticated to use American technology for sensitive, dual-use applications. It’s ludicrous that this technology is transformative for every field except for the military.

Dmitri Alperovitch: It doesn’t make sense. AI will transform everything. Even in civilian uses, do we want China to win in automotive, energy, and everything else? Because that’s what you’re enabling by selling chips to them. The primary concern is their military and intelligence services, but we are also in an economic competition. I would rather kneecap Chinese competitors to enable our own companies to succeed. Why would you do otherwise?

This is equivalent to selling supercomputers to the Soviet Union in the 1970s. No one even considered doing that. You could make the case that it would support Soviet agriculture and feed starving people, but no one said that because those same computers could be used for nuclear weapons testing and countless other military applications. There was no debate about it — it was understood to be a bad idea.

50 years later, we’re in a cold war. This is unbelievably shortsighted — putting profit above national security. Jensen Huang said if you’re a China hawk, you’re unpatriotic and un-American. I think selling supercomputing capabilities to the Chinese military is as unpatriotic and un-American as it gets.

Jordan Schneider: Jensen, if you’re listening, you’re invited to come on ChinaTalk anytime to make your case.

Dmitri, what’s telling is that the rest of the tech industry is finally pushing back. After months of staying quiet for fear of losing access to Nvidia chips, major players like Microsoft and AWS are supporting measures like the GAIN Act. The benefit of selling chips to China is mostly limited to Nvidia. U.S. hyperscalers and AI labs now face a powerful new competitor for limited chip manufacturing, driving up prices. The upside seems narrow, especially when Nvidia’s strongest argument — that the world, including China, will be locked into CUDA — seems far-fetched.

Dmitri Alperovitch: Nvidia’s argument is knowingly false. The GAIN Act is the ultimate ‘America First’ act. It stipulates that before chips are sold to countries of concern like China, we must ensure that U.S. demand is satisfied. American companies are first in line. How anyone could argue against this is beyond me.

The Act doesn’t say, “we’ll cut China off completely to ensure their military doesn’t get chips” — we’re saying, “let’s make sure American companies have priority.” It’s a no-brainer. I’ve talked to hyperscalers who are supportive of this act, and even other chip companies are saying they agree with the concept. The fight wasn’t about the details — the fight was a push for no restrictions on sales to China, which is unbelievable.

Jensen’s argument that the U.S. wants to make China addicted to the American tech stack is ridiculous. There is no addiction — chips aren’t cocaine. You can see this today with every single hyperscaler — Google, Amazon, Microsoft with its Maia chip, and now Meta with its own custom chips — all saying they are moving off CUDA. Many already are.

The top two frontier models, Claude and Gemini, were reportedly trained on Amazon’s Trainium and Google’s TPUs, respectively. There aren’t enough chips to go around, and for cost and strategic reasons, pretty much every frontier company is now using a multi-chip architecture — CUDA, Trainium, TPUs, and others. There is no addiction. Companies were able to make that switch in months, it’s easy — this is software and APIs. You can give AI one API and tell it to rewrite it in the form of another. It’s a trivial task.

Now we’re selling China H200s. This is probably the start of a broader concession on Blackwell, and then Rubin. Jensen won’t stop at the H200 — he will want to sell everything. The Chinese want to receive the latest and greatest chips, not only the Hopper generation. We’re going to sell them these chips, and they’re going to build competitive models. DeepSeek, Qwen, and Kimi are already good — they’re at most 12 months behind. They will quickly catch up and become leading models.

China will keep investing in Huawei because the Chinese are not stupid. Jensen says that if we don’t sell them chips, they’ll invest in their own, like Huawei’s Ascend chips. They’re doing that anyway. Xi Jinping is going to demand it, which is why you’re seeing China’s response that they will restrict the importation of H200s to ensure there is still domestic demand for Huawei chips.

Huawei’s Ascend chips will eventually catch up, and Chinese companies — supposedly “addicted” to the American AI stack — will switch over in days or weeks. What will we have achieved? We will have relinquished our lead in frontier AI models, and eventually, they’ll have chips that replace Nvidia’s. It is myopic and stupid for Nvidia’s own business model. They are focused on the next quarter and the next year versus a couple of years from now when China dominates both chips and frontier models.

Jordan Schneider: If this goes through, and tens of billions of dollars worth of chips are exported to China, and the future you portend comes true, will there will be a political price to pay? This was a major talking point for Trump on his campaign — “Winning the AI race” and “American AI dominance”. A year or 18 months from now, if China is releasing crazy new AI-powered technologies that were all trained on Nvidia chips, that will be a tricky political dance. Nice calls from Jensen won’t be enough to smooth that over.

Dmitri Alperovitch: We are already there. Almost a year ago, there was a brouhaha over the release of DeepSeek. The surprise was unwarranted: it shouldn’t have shocked anyone paying close attention. But people reacted with, “Oh my God, the Chinese are catching up.” Of course they are. Deepseek was built on H100 chips, which, until recently, were not restricted. There will be another DeepSeek moment, but worse. DeepSeek was good, but it was still behind frontier models. The next models will be better.

Sam Altman is in panic mode over Gemini 3 because its capabilities eclipse his models. This will happen to all American frontier models and to the country more broadly. The Chinese will crush us with cheaper power, tons of researchers, and massive state subsidies. The one thing they were missing — compute — will now flow into China.

Jordan Schneider: The Financial Times reported Chinese companies were training models in Malaysia or Singapore. That’s not ideal and not as efficient as AliCloud’s operations in China. There, they can rapidly deploy numerous H100s while benefiting from straightforward communication, a reliable power grid, and lower energy costs.

Dmitri Alperovitch: We should have been cracking down on H100 access in Malaysia and elsewhere. Chips shipped directly to China will be prioritized for high-side intelligence and military networks. Chinese agencies can’t use public clouds in Malaysia for their classified data. But now they can grab those chips from private companies in China and prioritize them for military purposes, as they do with everything else.

Jordan Schneider: That seems like the most salient reason China would want the chips inside the country. Training models in Malaysia is annoying, but only 10%-annoying. There are also data privacy restrictions, which they can get around if they’re serving domestic consumers in China. What do they want complete control of their chips for? The sensitive stuff that they would never trust a random Singaporean cutout to do for you.

Dmitri Alperovitch: The U.S. government cannot get enough chips. Agencies have told me they are compute-dependent for inference and cannot get enough chips. Now we’re shipping part of that limited supply to China. How does that make sense?

Jordan Schneider: Let’s flip this around.

Dmitri Alperovitch: One more point. The H200 is from the Hopper generation, not the latest Blackwell generation, but it has High-Bandwidth Memory (HBM). We have a current ban on the export of HBM to China. The H200 decision calls HBM protections into question, as the technology is already being exported on Nvidia chips.

We may see a cascading failure of export controls. I am hearing of discussions about relaxing export controls on semiconductor manufacturing equipment, which would make it easier for Huawei to manufacture Ascend chips in China. I hope that doesn’t happen, but there are people in the administration pushing for it.

Jordan Schneider: A year ago, the administration was being pressured to restrict chip technology to China. First there was the H20 situation, then the Laura Loomer saga and teh “twilight of the China hawks.” Lawmakers Vasant, Greer, and Rubio even intervened right before the Xi Jinping meeting to urge against concessions. Now, only a month later, this policy has been enacted without any clear reciprocal action from China other than continued soybean purchases.

Dmitri Alperovitch: I don’t know.

Jordan Schneider: To be determined. The main thing they’ve done recently is bully Japan. That’s the only big new development. And now we’re deciding to throw this other carrot into the mix. It’s weird.

Dmitri Alperovitch: The crazy thing is that China isn’t even asking for this. It didn’t come up in the Trump-Xi meeting. This is a concession to Jensen Huang, enabling Nvidia to make money at the expense of U.S. national security. I could understand it if this were a trade to get something we desperately want from China, like rare earths or a commitment not to invade Taiwan — though they would never do that. But it’s not. We are getting nothing for it. It is a favor to Jensen, to China, and to the PLA.

Jordan Schneider: It’s not even a big favor to my 401(k) — it only went up by two and a half percent. Come on.

Dmitri Alperovitch: Nvidia is in trouble because its U.S. market is going to shrink. Its primary customers, all the major hyperscalers, are building their own chips and want to move off of Nvidia’s platform. It’s desperately looking for another market, in China and the Middle East. That is why the company is pushing so hard for these export controls to be lifted. Jensen probably sees this is an existential problem.

Jordan Schneider: Dmitri, I appreciate your energy. I am so tired of these guys. I have to give Jensen credit for his stamina in making those calls and fighting through this. He has delivered twice now.

Dmitri Alperovitch: And he killed the GAIN Act.

Jordan Schneider: The man’s on a roll — he’s scored a touchdown.

Dmitri Alperovitch: And by the way, he’s not only going after the China hawks. The entire industry — from the hyperscalers to other chip companies — is on the other side of the ledger. He’s single-handedly beating everyone in this town. It is astonishing.

King Jensen

Jordan Schneider: Last year I asked you why more rich people don’t invest their time and energy to shape political outcomes. The thesis was that if you put the time and work in, you can get results. This is Exhibit A for CEOs trying to push through initiatives that may not have polled well initially. If you put in enough legwork and time on the phones, you can make things happen.

Dmitri Alperovitch: You have to give him kudos — he’s done incredibly well at the influence game here in D.C. He is putting in the time, meeting with anyone. He even said he’ll meet with Elizabeth Warren, one of his chief critics on the Democratic side. He’s calling the President almost daily, it seems. He got this done by badgering the President, repeating, “Get me my chips, get me my chips, get me my chips.” Donald Trump finally said, “Fine, here you go.”

Jordan Schneider: This development suggests the administration dismisses both the national security and the economic arguments for restricting this technology. It ignores the reality that these chips are vital in a strategic military competition.

Economically, it also overlooks the fact that strengthening Chinese competitors will harm American industry for decades. We should be consolidating the technology that drives productivity, not ceding it to a rival.

Dmitri Alperovitch: I don’t agree. The majority of this administration is opposed to this decision and does believe we are in a strategic competition with China. Call it a cold war. I know people in the administration agree. The president was convinced that selling China American AI stack is good for American business, and that Chinese firms will be addicted to it. But it’s a nonsensical argument. Jensen lied, because there is no addiction to the stack — it’s easy to move off of it. Unfortunately, he has been able to carry the day for now.

Jordan Schneider: This isn’t selling the “stack.” Selling the stack would be Nvidia chips run by AWS or Google, running Western models. This is selling the lowest level of the stack. I guess if the semiconductor manufacturing equipment (SME) relaxations come true, we’ll be selling the two lowest levels.

Dmitri Alperovitch: This is the equivalent of selling Ford cars to China in the hope China will be “addicted” and not prefer any other car. It is stupid on its face.

Jordan Schneider: It’s not even selling the Ford car — it’s selling the axles.

Dmitri Alperovitch: That’s all it is. There are huge problems with this decision. First, this is enabling the Chinese military and intelligence services, which are adversaries we could one day be at war with. The DoD is planning for a fight with China and stressing the need to overmatch its capabilities. Second, it puts Chinese firms on equal footing with American firms. Why would we do that? It hurts American companies and the American economy.

Jensen’s argument against export controls is inconsistent with his own business practices. He claims controls only encourage strategic competitors to innovate. By that logic, he should open-source his proprietary CUDA framework to AMD, because God forbid they develop a superior alternative. He doesn’t practice what he preaches. He is protecting his technology with patents and trade secrets, like any other company. Yet, he insists the U.S. should use a different strategy at a national level. It’s insanity.

Demand for chips in the U.S. already outstrips supply. Diverting this limited resource to a strategic military and economic competitor is a self-defeating act — we are actively surrendering the Cold War. I’m not an “AI doomer” — this technology is profoundly important for economic and military power. That is why there is no valid argument for helping your main rival develop it.

Jordan Schneider: Hey, White House. Hey, Nvidia. If you want to come on ChinaTalk and make those arguments, we could hash it out here.

Maybe we’ll be saved by the Ministry of State Security, who convince themselves that this is a crazy CIA plot to backdoor hack the PLA. It’s a longshot.

An Institute for Progress chart shows the U.S. and its allies currently possess a large compute advantage over China, roughly a 13-to-1 ratio. Selling large volumes of chips to China could drastically change this balance.

The main question is how Huawei’s domestic production compares to Nvidia’s global output from its fabs. If we withhold advanced equipment and AI chips from China, we can confidently expect a continued U.S. advantage. If these sales go through, it’s unclear who will lead in compute power in next 5 to 15 years.

Dmitri Alperovitch: It will be China, because they’re going to subsidize the hell out of this and we won’t.

It’s not over. Capital Hill is upset about this. Don’t count out Congress, the GAIN Act isn’t dead yet. There will be a fight to prioritize chips for American companies and to see what restrictions are possible — maybe export control reviews by Congress. There are bills floating around.

Also, Donald Trump often changes his mind. Others may convince him to revert this decision. The good thing about Donald Trump is that you’re never done. Whatever happens today can be undone tomorrow, and we need to take advantage of that.

Jordan Schneider: That’s the great irony in all of this. Given the political hesitancy on both side of the aisle and the possibility of Trump changing his mind, Alibaba, Tencent, or ByteDance are unlikely to bet their firms’ futures on Nvidia chips. This is going to be a political football, and one Truth Social post won’t end it. The strategy of “addicting Chinese firms” over the long term — setting aside Beijing’s own goal to indigenize chip production— won’t work.

Dmitri Alperovitch: Beyond politics, this strategy fails for basic business reasons.

China won’t get enough chips. You have Jensen acting as king, allocating a scarce supply of Nvidia chips to hyperscalers and now Chinese customers. Since there isn’t enough to go around, that scarcity forces them to rely on other chips.

No one wants to pay the “Nvidia tax” or be completely dependent on a single monopolistic supplier. Everyone wants to diversify, which is why you see them all building their architectures on multi-chip designs. Committing 100% to CUDA, politics aside, makes no commercial sense.

The Magic of AI

Jordan Schneider: Let’s close on some vibe-coding. I can’t be too depressed going into the holidays. Dmitri, I hear you’ve been having some fun with Opus 4.5 recently. What’s it done for you?

Dmitri Alperovitch: It’s magic. Anyone with a bachelor’s degree, not even in a technical field, can be a software engineer within three years, if not sooner. It is so easy to develop applications. I’ve built two mobile apps in the last month and a web app for personal use. Opus 4.5 is magic. I built a mobile app yesterday in 15 minutes, and most of that time was spent on setup, authorizing it on the Apple Store, and configuring my device. The capability is incredible, and it’s improving everyday.

This is the innovation we have to look forward to, and we want to make sure our American companies, our government, and our citizens are the primary beneficiaries. We want American frontier companies to be the best, and then we can restrict these models from actors we don’t want to have access.

I’m on the board of a number of companies, and I’m telling them all to start measuring their engineers on their use of AI in development tasks. Anyone who isn’t using AI should be considered for a performance improvement plan (PIP). This is the next hammer. It’s like when hammers were discovered tens of thousands of years ago — whoever didn’t use them fell behind. This is an unbelievable productivity tool.

One of my companies has a software engineering team developing their products. They’re also pulling people from other departments, like security, to help build the next module in Claude or other models. These teams are creating prototypes, and even production-ready versions. It’s unbelievable how you’re able to raise the productivity of everyone, not just software engineers.

Jordan Schneider: I want to say the same for analysts, think tankers, Hill staffers, and folks in the executive branch. It is a superpower. We were having a debate about whether Huawei can backfill Nvidia and what the ratio of chips would be. It took me 45 minutes to build an entire data visualization with sliders for different assumptions. How much HBM will China get? How tight will the export controls be? How much will they improve using DUV? How far behind will Huawei’s chips be compared to Nvidia’s?

Beyond the fun personal applications, it’s the “bicycle for the mind” aspect that people should experience, especially for thinking through policy problems. If you’re wrestling with a knotty issue that has numbers, contingencies, or second-order effects that are hard to hold in your head, ask Claude to help you visualize it or see the other side of the argument.

The hallucination issue is almost gone. You still need to fact-check the details and trust your gut if something seems off, but the improvement has been dramatic.

Dmitri Alperovitch: It depends on what you’re using it for. At some level, it’s garbage in, garbage out. If you’re training a model on Reddit and asking about something very esoteric, you’re not going to get a good answer.

Jordan Schneider: You are doing yourself a disservice if you haven’t spent time with these models. Try to integrate them into your day job. You should be hanging out on Cursor and Claude, trying to build little tools and apps to make your workflow easier or allow you to do new things.

Dmitri Alperovitch: Building apps was nostalgic for me. It brought back the emotions I felt as a kid in the 1980s when I learned programming. It was an amazing feel coding your first “Hello, World!” program or, in my case, a simple game in QBasic. The magic of seeing it run was a special feeling, and you felt so proud and accomplished.

This took me back. It made me think, “Oh my God, this is magic.” In the ‘80s and ‘90s, you had to have technical expertise and learn a programming language. You still need some technical skills today, particularly when you’re debugging or if you don’t understand how Swift works or how to deploy iOS apps. But all of that is going away.

Jordan Schneider: It’s going away.

Dmitri Alperovitch: The accessibility of this technology changing everything. For years, we thought only nerds could access the magic of programming. Now, everyone can, and that is going to revolutionize everything. The interesting thing about AI is not that it’s going to make tasks easier and faster, but that it’s going to make other things that you would never, ever do before accessible.

The cost of software engineering iwill drop to zero. Everyone will be building dozens of apps — for their grocery list, for managing their kids’ schedules, whatever it may be — because it’s so easy. You can custom build something that would be useful only to you, with no commercial value. Even for coders, we wouldn’t spend our time building those apps it was a lot of effort. Now, that effort is gone.

Jordan Schneider: The activation energy for doing a side project has dropped to zero. What I’m excited to see created, Dmitri, is the “senior policy official simulator.” That’s a classic nerdy ChinaTalk idea.

Dmitri Alperovitch: So nerdy.

Jordan Schneider: But you read all these memoirs from government officials. Jake Sullivan said the one thing you can’t experience beforehand is being in a crisis. You can have a Tim Geithner level — all of a sudden it’s 2009, and it’s not like you’ve lived through a financial crisis before.

Having a visceral experience — a VR Situation Room meeting, a VR flight on the plane with the president trying to convince him not to sell chips to China — getting reps in those high-stakes political, personal, and commercial situations could be transformative. It doesn’t have to be for politics and national security. We haven’t had a nuclear crisis in a long time.

Having the deeper, emergent human capabilities that AI simulations of these events can provide seems like a big upside for human competence when dealing with crises in the future. I’m excited about it. Rockstar Games, if you’re out there, give me a call. We can do some cool stuff together.

Dmitri, always a pleasure. Thank you so much for being a part of ChinaTalk.

Dmitri Alperovitch: Thanks for having me, Jordan.

Mood Music

大家好，本期放学以后信号塔由西班牙的霸王花木兰轮值。现在我正在塞维利亚的民宿沙发里窝着写这篇newsletter。

我又出门游荡了。

先是莫不谷发起了去Alicante吃海鲜过春天的活动，我和莫不谷，还有荷兰朋友小鱼儿以及芬兰的粽子一起到阿利坎特Alicante集合，度过了短暂却精彩纷呈的三天。

由于芬兰飞西班牙机票太贵，粽子来一趟不容易，所以在莫不谷的提议下，结束Alicante行程后，粽子又和我一块背包开启了西班牙南部的游荡之旅，先是去海鲜饭发源地瓦伦西亚Valencia，再飞去美丽的南部城市塞维利亚Seville，接着坐大巴到达热门旅行城市马拉加Malaga，最后粽子从这里结束游荡，飞回芬兰。

这一路的行程，粽子像是海绵一样遇水膨胀，疯狂储蓄西班牙的阳光，这样在芬兰夜长日少的时间里慢慢使用，明年四五月份春天也就没有那么遥远了。今天粽子还在塞维利亚的街上边走边唱，我的快乐，不会来了~因为一想到西班牙游荡行将结束，她便心生不舍。不过她也计划以后每年冬天都来一趟西班牙，因为阳光太好了！莫不谷也早早计划了2026年的冬日游荡行程：到西班牙的最南端——科尔多瓦+加地斯+龙达+直布罗陀 （对面就是非洲） 。

而我则是终于有机会来南部游荡，特别是莫不谷曾在四月去过腿部干裂到出血却仍止不住夸赞的塞维利亚Seville。

落地第一天上午，在灿烂到背部发热的阳光下，我背包徒步走到了西班牙广场，这个简直是我全世界游荡以来遇到的最美丽的广场，评为我的世界第一广场也不为过。恰逢广场里舞蹈正在配合热情的音乐演奏，旁边烤板栗的香气和烟雾在人群头上聚集又随风飘去，五彩缤纷的泡泡正在被人吹出，又被好奇的孩子们追逐试图拍散，阳光洒在粉橙色的广场建筑和缤纷美丽像是“青花瓷”又像是“唐三彩”地瓷砖上，广场正中央喷泉正在以蓬勃的状态喷洒水滴，细碎的水雾遇上阳光折出一抹七色彩虹，当人的视觉，听觉，嗅觉被偶然聚合的景象充分打开，热情，阳光，梦幻，快乐，兴奋，这些词便在我的脑海里陆续飘过。人对一个城市的第一印象就此形成，美丽。

塞维利亚有一款橙子味道的香水，还有一个美丽的名字，塞维利亚的空气。因为塞维利亚路边全是橙子树，春天是白色橙花绽放的季节，满城飘香，冬天是橙子结果的季节，也是橙子汁水最足，味道最好，价格便宜的季节，来到这里，见过冬日街边满树橙子的人，谁又能拒绝这款塞维利亚的空气呢？

我走进了莫不谷保存在Seville googlelist里的一家小众香水店，Naturally, aromas of Seville，店员在我的手腕上试了orange bloosm的香水后，我便忍不住把这款美丽的味道买下，出门走在大街上，整个人被成熟的橙子圈起来一个味道的结界，感觉自己变成发现美味食物疯狂闻嗅的狗，抑亦或者变成猫薄荷上头的猫，不停用鼻子闻着喷过香水的手腕，再使劲嗅着空气里的留香，难以自拔。

这次南部游荡是和粽子一起，Alicante游荡是和莫不谷还有荷兰朋友小鱼儿一起。我在香港，波兰，英国等地体验过solotrip，一个人游荡的好处是自在，随性，方便，体验自己独自和世界交手。而我更多地是和朋友一起游荡，既有和莫路狂花的多次全球游荡，也有和女性朋友们一起的集体游荡。与人一起便有相处的问题，特别是在游荡过程中，价值观，消费观，性格特点，生活习惯，个人胃口，出行安排等等是否能够合得来，都会影响游荡体验。

而另一方面，和朋友们结伴游荡的好处也很明显。《拼团人生》这本书里说，分享快乐，快乐会加倍；分享悲伤，悲伤会减半。在Alicante阿利坎特和女性朋友们一起出门游荡，去美丽的地中海边徒步，大快朵颐人均19欧元海鲜饕餮自助，晚上一起投影看“快乐小偷”英文脱口秀，又在第二日一起去中央市场采购新鲜牛肉和海鲜，共同制作美味的贵州酸汤牛肉海鲜火锅，吃完再一起打扫收拾。

人多热闹也多，能够一起创造的记忆也多，感觉和相处愉快的朋友呆在一起，生活也变得丰富和有意思起来。

我不是群居动物，我常常独居，不爱聊天，对人没有好奇心，喜欢独自行动，遇到事了还喜欢躲在蜗牛壳里避免被发现。可与志同道合的朋友短暂共居的我会感到比平时多一点的安心，这个感觉挺奇妙。就像是自己独自玩游戏，过程很投入很开心，结束后总还是有些虚无。但是和人一起玩游戏感受就会有些不同。

前段时间莫不谷发起了“你画我猜”欧洲女子联赛，和芬兰的粽子，瑞士的Ruya，Ruya的意大利米兰朋友Max，还有荷兰的朋友茶茶分别玩了三场在线游戏，每次都玩两三个小时不过瘾，灵魂画手的我画的A4纸都要冒火星子，而玩游戏时，我的胜负欲，集中力，想象力和创造力也被高度调动起来，甚至前一天通宵看网文《祝姑娘今天掉坑了吗》睡不够，也丝毫不影响玩游戏。

也因为如此，莫不谷的每次游戏提议我都很心动，游荡提议同样心动。

最近沉迷看《祝姑娘今天掉坑了吗》，网文里的大部分角色无一不被祝姑娘折服，相信跟着祝姑娘不愁人生前路。网文外的我，也忍不住为祝姑娘折服，不自觉代入想跟着她干事业的人物角色。又总在阅读文字的时候想到莫不谷，跟着小祝大人不愁前路，跟着莫不谷则是不愁美食和职业。

这次在Alicante游荡，就跟着莫不谷一起体验了新的美食，还经历了神奇的际遇。还没来西班牙前，莫不谷就一直沉迷熟成鱼，咔咔学习熟成知识，时不时和我分享美食视频，还建议我去寿司店学习一下，以后开餐厅就可以负责活缔杀鱼。所以她在确定Alicante游荡机票的当天，就找好了熟成牛肉的餐厅，因为熟成鱼要去伦敦和巴黎才能吃到而且价格高昂。

没想到在Alicante一家中超店铺里的包子铺吃早餐时，莫不谷又在念叨这件事，谁能想到小小包子铺的老板是全世界钓鱼的爱好者，最近一次钓过300多公斤，价值一万多欧元约合人民币十几万的金枪鱼，而恰好他昨天刚钓的，做了活鱼取缔，还没来得及送给朋友的金枪鱼亚种就在包子铺的水桶里。更难想到的是这个山东青岛老板0桢起手，二话不说便把鱼拿出来现场处理做刺身，同时搭配了酱油和芥末的日式吃法，柠檬和白糖的泰式吃法，还用打火机将柠檬和白糖在三文鱼上烤制一下，最后将整条鱼全部免费送给我们品尝！

这是任谁怎么想都想不出来的奇遇，但和有着强烈渴望和心愿的莫不谷一起，感觉什么奇迹都有可能发生。

（莫不谷在游荡者网站分享的aha moment）

另一个惊奇的小故事是，我们前一天晚上在Alicante中央市场买了新鲜便宜，只要2欧一个的地中海蓝蟹，简单水煮就可以吃到清甜可口的蟹肉。因为吃不过瘾，第二天莫不谷又要飞回荷兰，飞行当天我又去中央市场给莫不谷带了两个螃蟹。由于生螃蟹无法带上飞机，莫不谷便提出一个在我看来很难想到，想到也做不到的方法，先去包子铺吃早餐，到时候请包子铺老板帮忙煮螃蟹。

我是真没想到这事能成。为避免被拒绝的尴尬我还提议要不要以支付加工费的方式试试，说不定老板能同意在包子铺帮忙煮螃蟹。结果吃完早餐消费完毕的莫不谷和老板开口说明情况，希望老板能帮忙煮一下，勇猛地开口不仅成功煮上了地中海螃蟹，还由此聊到她心心念念的熟成鱼，接着就是意外惊喜地吃上了珍贵的活鱼刺身。

跟着这个小故事的后续是，莫不谷在飞机上突然太饿了，干了一件极其疯狂的事，在飞机上把两个螃蟹啃干净了。然而不仅没有人投诉反馈，对面荷兰人还热心借给她湿纸巾，空少还过来帮忙收拾垃圾。对我来说，真是惊奇，震惊和佩服打个包裹在一起，一波又一波来袭。

而另一个印象深刻难以忘怀的游荡奇遇是，这次我们居然在Alicante遇到了双彩虹！今年春天我和莫不谷一起来Alicante爬山时，风景超级美丽，由于阳光太好洒在巴拉巴拉城堡城墙时，有一种不必吃苦人就来到了埃及的感受。

这次11月来Alicante再爬巴拉巴拉城堡，冬天植物没那么茂盛，天气也有些阴天，完全没有了埃及的感受，甚至爬山中途还突然下起了小雨，可就在我给粽子拍照时，一抬头看到了两轮巨大的彩虹悬挂空中，一头连接Alicante这座城市，一头直直插入蔚蓝清澈的地中海，仿佛这难得的双霓虹是从神秘的海里生长出来的，我们惊奇地喊出声来，山上的人们纷纷拿起手机抬头看向美丽的天空。

斯景双霓虹，遇上方知有。游荡路上的奇迹，是出了门看到，吃到，体验到，真的有可能发生，甚至一定会发生在有着强烈渴望和热情的人身上。我不像莫不谷那样对美食，创作有激情，热情和渴望。虽不能至，心向往之。所以我先出个门再说。

最后分享一些本次游荡的图片！

成为放学以后Newsletter月度会员，可以解锁既往所有付费内容，解锁完记得在权益期及时查看所有付费内容，以最大化享受权益。如下月不再继续付费订阅，也记得及时解除，以防发生计划外扣费；爱发电支持购买单期付费播客或文章。大家可根据自身情况选择最适合的方式，苹果用户请不要下载appstore的爱发电app，是诈骗。

放学以后爱发电“电铺”：https://afdian.com/a/afterschool?tab=shop

《创作者手册：从播客开始说起》（小册子）系列https://afdian.com/item/ffcd59481b9411ee882652540025c377

run&rebel系列1《朋友们，Run and Rebel：快逃以及反抗！》https://afdian.com/item/2b3a33acfd3311ecb4d852540025c377

run&rebel系列2《在这个时代，做个反派》https://afdian.com/item/b9c74240bcff11ed86fe5254001e7c00

run&rebel系列3《爹和爹味，吐槽大会》https://afdian.com/item/6529d622092011ee8a1352540025c377

run&rebel系列4《活在历史的垃圾时间，我们如何度过时代的乱纪元？》https://afdian.com/item/90682ea4c68611ef8e645254001e7c00

run&rebel系列5《让我们不吐不快：各行各业，各个工种，各色牛马，吐槽齐发》https://afdian.com/item/87b95f1ac32111f0b10552540025c377

放学以后《莫路狂花今夜不设防：人如何不糊弄和痛恨自己，并找到自己的渴望呢？》https://afdian.com/item/e4b68686a67911ef8f2f5254001e7c00

放学以后《莫路狂花2：如何对自己充满爱意和敬意，免于混乱逃避低活力？》https://afdian.com/item/3572eaba3a6d11f0ac9052540025c377

放学以后《终身学习1：学会面对真问题，不逃避，下决心和谈分离》https://afdian.com/item/e96a78d4619c11f09e8552540025c377

游荡者平台：www.youdangzhe.com 或者www.youdangzhewander.com

Tony Stark and Justin Mc return for Second Breakfast. In Part I, we break down the Trump administration’s new National Security Strategy (NSS).

Today, our conversation covers…

• What a National Security Strategy is, and why they matter,

• Controversial new inclusions in Trump’s NSS, including on Taiwan policy and the “reinvigoration of American spiritual and cultural health,”

• How to reconcile the document’s ambitious vision for deterrence with the reality of Trump’s China policy,

• The mixed signals this NSS sends to U.S. allies,

• What Buffalo Wild Wings can teach us about competition with China.

Listen now on iTunes, Spotify, or your favorite podcast app.

Ends, Means, and One China

Jordan Schneider: Tony, give us the 101 on what a National Security Strategy is, and then we’re all going to go around and say one nice thing about it.

Tony Stark: There are three major U.S. government national security strategy documents. The first is the National Military Strategy, which applies to the uniformed services but is rarely noticed outside the Joint Staff.

Next is the National Defense Strategy (NDS), which is the Pentagon’s primary strategic document. It’s the one most people in the field care about because it’s a Cabinet-level document, even if it isn’t overtly political. Legally, a new NDS is required every four years, and developing a new NDS takes 6 to 18 months. New administrations are given a little extra time — about a year and a half — to publish their first one.

The NDS is written at the “action officer” level, which includes General Schedule (GS) employees, field-grade officers, contractors, and think tank experts. Then it is passed up to the Deputy Assistant Secretary level in the Office of the Secretary of Defense (OSD) — their equivalents are three-star generals — and then to the commands, the undersecretaries, and so on.

Finally, there’s the National Security Strategy (NSS), which is historically the most political of the documents because it comes out of the White House, not the Pentagon. The NSS is a guiding vision of the administration’s goals and incorporates all elements of national power. Historically, this is also the blandest document — its wide scope reads more as a political statement than a defense plan. The new Trump administration just released its first NSS. While the NDS has been ready for a while, they were likely waiting to publish the NSS first.

At 29 pages, the new NSS is the right length for a public national strategy document. There are usually non-public, classified annexes and other materials.

Justin McIntosh: The document correctly focuses on economic re-industrialization and re-energizing the defense industrial base — issues we’ve previously discussed. It puts those ideas forward in its “answers” section. But…

Jordan Schneider: No “buts.”

Justin McIntosh: Okay! Yes, that’s where the focus should be.

Jordan Schneider: The straightforward questions in the document are nice. The Q&A rhythm is interesting and provocative. It’s focused. There’s a section of questions like, “What should the U.S. want overall?” and “What does the U.S. want from the world?” There’s no artifice about how transactional it’s going to be — what you see is what you get.

Tony Stark: If I were framing a strategy document for the American people, this is how I would structure it. A clear layout saying, “This is what we want. This is why we have a strategy. What are the ends, ways, and means? What does that mean?” It’s written in a clear, accessible way, without many buzzwords. Although what replaced the buzzwords wasn’t great.

Jordan Schneider: Avoiding policy jargon in this document seems to have been a conscious choice.

Justin McIntosh: But it lacks nuanced, impartial language and contains statements that our adversaries will exploit. A comment on the necessity of securing borders said that any sovereign nation has the right to control them. The PRC and Russia can easily seize on a statement like that. This is a kind of language previous administrations have avoided, because they didn’t want a quote interpreted as agreeing with the Chinese or Russian position.

Tony Stark: The document does not change U.S. policy towards Taiwan. If anyone tells you it does, they are wrong. However, it does give the PRC political and legal ammunition. They can now say, “But you said you wouldn’t interfere in the internal affairs of others,” pointing to our supposed principles of non-interventionism.

The document also says we do have to intervene sometimes. This amounts to talking out of both sides of your mouth — we reserve the right to do whatever we want. The “flexible realism” section is a fancy way of saying we’ll do whatever is convenient. Historically, that has been U.S. foreign policy in practice, but that doesn’t mean it’s what we should aspire to.

Justin McIntosh: I don’t have a problem with them laying out the “ends, ways, and means” discussion up front, but it has limitations. That linear framework is well-suited to military decision-making, but a national strategy needs to be more pragmatic and flexible. At the national level, you control all the resources. You can marshal all those resources toward any goal that is deemed important. That makes the “ends, ways, and means” calculation irrelevant because you will find a way to make it happen.

Jordan Schneider: The Trump administration’s focus on “ends, ways, and means” raises the question — how weak do they think the U.S. really is?

Reducing the U.S.’s power to an “ends, ways, and means” calculation only works in military contexts — counting ships and battalions to see how many wars you can fight. The U.S.’s power to achieve economic and national security ends is elastic. The means to those ends can grow dramatically when the president builds a consensus around them — once the nation decides something must be done, it finds the capacity to do it.

It’s a mistake to define goals downward because those goals inevitably change. Consider the border — the Biden administration didn’t prioritize the issue and struggled to find the means. The Trump administration’s intense focus on the border unlocked congressional funding and operational capacity. The resources didn’t appear from nowhere — the will to use them did. This dynamic applies globally. To believe the U.S. cannot act because it lacks on-hand capabilities is a severely limited way of thinking about our power to shape events.

Mixed Signals

Tony Stark: The document’s focus on military and economic power isn’t unique, but its goals do not align with a realistic budget. It calls for both bolstering deterrence in the Indo-Pacific and shifting our entire global military posture to the Western Pacific, which would drain resources from Europe and Latin America. We have to assume this will happen.

This creates deep concern for our allies, but that matters for the U.S. too. The Germans will be wildly pissed about how they are described in the document. Asian allies are told to “do more,” a demand that ignores their significant recent efforts. Getting allies to increase defense contributions was an accomplishment of the first Trump administration that continued under Biden. The call to “do more” is now an outdated talking point — they are doing more. Japan is considering exporting weapons for the first time.

Justin McIntosh: Worse still, when allies make the kinds of statements the U.S. wants — like Sanae Takaichi declaring a PLA incursion into Taiwan a national security threat to Japan — the administration’s response is silence. Based on the reporting of Xi and Trump’s call, it appears the U.S. did not affirm that position. Instead of backing Japan’s strong stance, the message was to “calm it down.”

The Trump administration is sending mixed signals. Does it want allies to spend more on defense, develop a stronger defense mindset, and care more about their own security, or not?

Jordan Schneider: Let’s do some reading from the scripture here.

“A favorable conventional military balance remains an essential component of strategic competition. There is rightly much focus on Taiwan, partly because of Taiwan’s dominance of semiconductor production, but mostly because Taiwan provides direct access to the second island chain and splits Northeast and Southeast Asia into two distinct theaters. Hence, preventing a conflict over Taiwan, ideally by preserving military overmatch, is a priority. We will also maintain our long-standing declaratory policy on Taiwan, meaning that the United States did not support any unilateral changes to the status quo in the Taiwan Strait.”

From that, it sounds like a good idea for Japan to make its role in deterrence transparent. How seriously should we take any of these documents?

Tony Stark: I wish Eric were here for another briefcase-carrier rant. In the 2010s, a gripe of mine was hearing mainstream national security people, the ones in the know, say strategy documents don’t matter. That is a clear indicator they either haven’t written a good strategy document or haven’t marshalled the resources and people to execute it. I’ve occasionally had to metaphorically beat somebody over the head with a strategy document.

One problem is that people don’t read strategy documents. I have been in meetings with theater-level commands who’ve asked me, “What are you quoting from?” And my response is, “The National Defense Strategy.” They’ll ask me to send it to them. It’s a public document.

Justin McIntosh: “No, no, we meant the classified annex, Tony. Obviously, we’ve read the public one.”

Tony Stark: “The super-secret one that wasn’t even fully distributed to your command.”

Justin McIntosh: The document doesn’t matter, and there isn’t a robust national security apparatus anymore — at least in this administration — it’s as if the President is the sole decision-maker. Trump has consolidated his counsel — it’s a smaller group than it was.

Another problem is that the strategy document’s promises are often the opposite of what the president himself has done. The strategy specifically addresses deterring propaganda aimed at Americans, clearly referencing China, and yet TikTok is still legal here.

When X turned on a filter showing where accounts came from, it revealed so-called Mongolian accounts weren’t Mongolian, and supposed Uyghur accounts were run from mainland China. Pro-MAGA accounts were operated from VPNs in India and China to target Americans. Where was the action on that propaganda? We kept TikTok, and no one has suggested the government force X to shut down foreign influence accounts. These goals are in the document, but the follow-through is missing.

Tony Stark: Every administration struggles with inconsistencies between its strategy and actions. That’s the nature of a democracy — it’s the nature of any government worldwide. This strategy document’s main issue is its unusual use of national security language. The strategy says the administration opposes disinformation, but what do they consider disinformation? There are direct quotes that frame concepts like “de-radicalization” and “protecting our democracy” as a fake guise — that inclusion is wild.

On foreign policy, the document critiques the U.S. for focusing too much on projecting “liberal ideology” into Africa — it’s unclear if that means big ‘L’ or small ‘l’ liberal. Let’s assume it’s both. The most stunning part is that the National Security Strategy of the United States explicitly frames the concept of “protecting our democracy” as a ruse. That is insane.

The parts of a strategy document that truly matter are the ones that diverge from the previous strategies. While I’ve critiqued previous strategies, this document is on another level.

Justin McIntosh: The large section on China is a good example. It would be great if the administration enacted many of the listed actions — I’d be all for it. The cognitive dissonance between the strategy document and the administration’s actions is troubling.

Jordan Schneider: Six months ago, the AI action plan included interesting language about new export controls on semiconductor manufacturing equipment. Those controls are paused because Stephen Miller’s current job is to avoid upsetting China. This directive came after a Chinese official was angered by a Financial Times article on Alibaba and the PLA. Stephen Miller’s Twitter banner is a picture of him shaking hands with Xi. This is hard to square with official strategy documents demanding military overmatch.

You can try to connect those dots and argue that the goal is to keep the economic relationship calm while we re-industrialize and build up our military. Okay, maybe. But that still doesn’t explain the U.S NSS includes sovereignty language seemingly copied and pasted from Putin’s playbook.

Traditional Values, Universal Wings

Tony Stark: The document is also very undergraduate. That is not a critique of the accessible language — I also try to write for a wider audience — but of the concepts themselves. If an undergraduate at the University of Texas at Austin were assigned the paper topic — what should a national security strategy be — this would be that paper.

Jordan Schneider: There are 14 bullet points where each sentence is about seven words long.

Tony Stark: What does this all mean? The language in the National Security Strategy should not shock anyone — it’s consistent with the administration’s usual rhetoric. What has changed is that this language is now the official guidance — it has leverage in bureaucratic fights. The influence may not be immediate, but it will be cumulative. The real test will be when the National Defense Strategy comes out. Someone who worked on it texted me last night and said, “Well, they set the bar low, so this is great for us.”

Justin McIntosh: They’re being pragmatic. What troubled me was the traditionalist language at the end.

“Finally, we want the restoration and reinvigoration of American spiritual and cultural health, without which long-term security is impossible. We want an America that cherishes its past glories and its heroes, and that looks forward to a new golden age. We want a people who are proud, happy, and optimistic that they will leave their country to the next generation better than they found it. We want a gainfully employed citizenry—with no one sitting on the sidelines—who take satisfaction from knowing that their work is essential to the prosperity of our nation and to the well-being of individuals and families. This cannot be accomplished without growing numbers of strong, traditional families that raise healthy children.”

Tony Stark: “We will use every means to protect our precious bodily fluids.”

Jordan Schneider: Wait, if you’re raising a disabled child, or if your child is sick with a fever, then you are not contributing to the restoration of American cultural and spiritual health? Wow.

Tony Stark: That is what RFK Jr. said — if your kid is sick, that’s not a good societal contribution.

Justin McIntosh: His miasmas are off, or whatever non-germ-theory medicine he peddles but doesn’t practice.

Tony Stark: The Midi-chlorians from Star Wars.

Justin McIntosh: That language is reminiscent of what you see from Putin and China’s family planning policies. It is the exact type of language that Xi and Putin use to justify pro-natalist policies and promote traditional families and traditional gender roles. Reading about the one-child policy in Dan Wang’s Breakneck is heartbreaking if you have children. It’s striking how similar the NSS’s language is to China’s early discussion of the one-child policy.

Tony Stark: In a reasonable time, there would be ten articles asking, “What does this mean? How is the government going to encourage people to have more kids?” Now, it’s something I don’t even want to read about.

After COVID-19, as the “China Rising” narrative was gaining prominence in 2021 and 2022, discussions began in national security circles about how the U.S. population is numerically outmatched. Although we are solving that problem with robotics, it was a talking point among traditionalists. They argued that the U.S. won the Cold War by embracing traditional values. That’s not how we won. We won thanks to Skunk Works and the Soviet Union’s economic mismanagement.

This argument has surfaced before in national security circles — it’s not a new phenomenon. The other common concern is protecting our food supply — I’m surprised it was not mentioned in the document. But, to quote a former coworker of mine, “We have Buffalo Wild Wings and the Chinese don’t. I think we’re okay.”

Jordan Schneider: That would be a great cultural export. Maybe that’s what the world needs.

Tony Stark: Are there Buffalo Wild Wings locations in Shanghai or Beijing?

Justin McIntosh: I’m sure there’s one in Taipei. [Note from Lily: Taiwan does not have a Buffalo Wild Wings, but it does have three Hooters locations.]

Tony Stark: Is the food different, or is it universal?

Justin McIntosh: It’s universal, but like McDonald’s in Japan, it’s better.

Tony Stark: Another American cultural victory. We don’t need to change anything.

Justin McIntosh: You can watch a baseball game while eating Buffalo Wild Wings in downtown Taipei.

Tony Stark: During COVID, my former American University professor, Justin Jacobs, uploaded all his lectures on Spotify — excellent lectures on the history of China and Japan. He has an episode about why baseball is played in Taiwan but not on the mainland. He discusses the Japanese occupation of Taiwan and the differences in Confucian culture and masculinity. Prof. Jacobs is an amazing resource for East Asian history.

Jordan Schneider: I asked Gemini what other regimes this resembles. It suggested Vichy France, Fascist Italy, and modern Hungary.

Justin McIntosh: I wonder what Grok would say…

亲爱的读者周末好，好久不见～

本期的主题是现代日本史。

方法

按照 “系统自学指南“思路，新领域入门的第一步是先找大学的课程大纲 syllabus。

Google 搜索或者直接询问 AI 助手，可以快速找到很多大学 Modern Japan / Modern Japanese History 课程大纲，比如：

• 佛罗里达大学 2022 年秋季学期课程 Modern Japan，由 James Gerien-Chen 陳慎仁教授讲授（点击下载PDF）

• 臺灣大學 “现代日本政治外交史” 课程纲要，由曾寶滿教授（东京大学历史学博士）讲授（点击查看网页）

UF 课程偏重社会、文化和生活史，臺大课程偏重政治外交；综合以上两份课纲所列书目，可得到一份精简的现代日本史入门书单。

戈登版现代日本史

英文：A Modern History of Japan, Oxford UP, 2019.

港版：《200年日本史》香港中文大学出版社，2014/2022。

简中：《现代日本史：从德川时代到21世纪》中信出版社，2017。

中文有港版和简中两个版本。

安德鲁·戈登任教于哈佛大学，曾担任哈佛大学日本研究所所长。这本当然是美国大学最常用的日本史教科书。

北冈伸一《日本政治史》

台版：《日本政治史》麥田出版，2018。

简中：《日本政治史》南京大学出版社，2014。

英文： Kitaoka Shinichi: The Political History of Modern Japan, Routledge 2018.

200 页写完 150 年日本政治史，气魄宏大。作者把国际关系和日本国内政治的互动讲解得十分清晰。

北冈伸一以大格局著称，影响力横贯政学两界。2010年代他曾是安倍晋三最信任的智囊。

戈登 + 北冈伸一，两本书的时间跨度均从德川幕府直达平成时代，第一本侧重经济社会生活史，第二本专攻政治史。我个人非常喜欢这套入门组合。

古川隆久《昭和史》

台版： 古川隆久 《昭和史》 玉山社，2019。

简中：《毁灭与重生：日本昭和时代》浙江人民出版社，2021。

中村政则 《日本战后史》

简中：中村政则 《日本战后史》中国人民大学出版社，2008。

“最后的讲座派”，鲜明的左翼立场。

五百旗头真 《战后日本外交史》

简中：五百旗头真 《战后日本外交史》世界知识出版社，2013。

《岩波新書‧日本近現代史》新书十卷

《岩波新書‧日本近現代史》香港中和出版社，2017。

1、《幕末與維新》井上勝生　著

2、《民權與憲法》牧原憲夫　著

3、《日清、日俄戰爭》原田敬一　著

4、《大正民主運動》成田龍一　著

5、《從滿州事變到日中戰爭》加藤陽子　著

6、《亞洲、太平洋戰爭》吉田裕　著

7、《佔領與改革》雨宮昭一　著

8、《高速增長》武田晴人　著

9、《後戰後社會》吉見俊哉　著

10、《應該如何認識日本近現代史》岩波新書編輯部編

写在最后

审核是道滤网。能被翻译成简中的日本现代史，大多左翼观点（比如岩波），这类书更容易通过中国的图书审查滤网。

因此如果只读中文书，相当于拿到一份撕掉了右半侧的地图。虽然不全，但也蛮精彩。

另外，纸上得来终觉浅，不妨去日本走一走。逃开漫天的时事喧嚣，读自己想读的书，行自己想走的路。

增广见闻，博大胸怀，没有比读书行路更好的办法了。

祝阅读愉快～

如果喜欢这个专栏，请推荐给家人朋友们订阅：

几天前，我从家里出发，一路向西，开车穿过得克萨斯。第一站目的地是得州边境的Big Bend——大弯国家公园。

开车路上，我第一次听到滨崎步这个名字，知道了她是一位日本女歌手。在此之前，我从未听过她的歌，对她一无所知。我听新闻说，她本来11月29日要在上海开演唱会，但中国政府在开演前两天勒令她取消演出。观众和票房当然都没了。但她没有放弃，没有抗议，而是带着乐队，面对空无一人的剧场，像正常演出一样，从头到尾唱完了整场。

那天上路以后，一直下雨，过了得州中部的丘陵地带，雨停了。路两边的树木越来越低矮，光秃秃的山丘连绵不断。得州西部仍然是人口稀少的蛮荒之地。在Sonora的加油站，一位开施工卡车的墨西哥人，左侧腰带上挎着一把1911。他可能是左撇子。人很友好，看我从车上下来，点头示意。得克萨斯是open carry state，可以把枪带在外面。但大部分人不会这样。我从来没有open carry过，都是别在外套下面，或者放在随身的包里。

得州西部硬朗、孤绝。它的本色就是这样，不是刻意做出来的。这就像一个人的性格。在车上听到滨崎步的故事，觉得她就像得克萨斯西部人一样，硬朗、孤绝。于是，我开始补课。在一个休息站，我在YouTube Music上找到她的歌，看了几首歌词的英文翻译，开始一首接一首地听。

听着听着，我好象领会到，为什么她能干出“空场演出”这种事。这是她的性格，是她的为人处世的方式，是她的做事风格。She just did what she had to do。

滨崎步显然不是那种糖果型女歌手。她的声音充满了瑕疵——你能听到她声带边缘的撕裂感，但这正是她歌声的迷人之处。她是用生命在撞击难以突破的极限。她的歌词写满了绝望、孤独，还有，更重要的是，绝望之后的重生。

她的歌声，她的空场演出，让我想到一个我崇尚的人生原则：“我们决定不了别人做什么，但我们能决定自己怎么做。”

她决定不了中国政府做什么，但她能决定自己怎么做——对着满场空座，完整地演出已经排演好的节目。有一股晦暗的力量，不让她的歌声被听到，她无法控制那股力量，但她决定自己一如既往，唱出自己的歌声。

“I do what I have to do.” 这句话说起来容易，做起来难。因为绝大多数人做事，是为了得到反馈——为了掌声，为了门票收入，为了面子。一旦这些没了，动力就熄火了。

但滨崎步展示了一种叫做“举重若轻”的境界。 面对一切到位之后却被取消演出这样沉重的打击，她像掸去衣服上的灰尘一样，表现出来的态度是：“随他去吧，我继续唱。”

这种轻盈，来自于强大的内心世界。她不需要把力量展示给谁看，因为她自己就是力量的源头。

跟滨崎步的举重若轻相比，中国政府的表现像群举轻若重的猴子。

一场商业演出，一位流行乐歌手。 在正常国家，这本来是一件微不足道的事。但在那个虚张声势的国度，这似乎变成了天塌下来的大事。党国如临大敌，发公文、拉警戒线、派警察维稳，好像如果不取消这场演唱会，政权明天就会崩溃一样。

这不禁让人想起土皇帝带给中文世界的那个梗——200斤。本来是个无能、自卑、虚弱的人，却吹嘘自己扛200斤麦子不换肩。不知道这是人还是骡子？

昨天登山，看到一队骡子驮着些铁皮盒子下山，看着也没有200斤。

那个自称200斤的，连100斤也扛不起来，所以他才看什么都觉得重。因为他内心虚弱，看什么都重得要命，看个乒乓球，都当成重磅炸弹，如临大敌。

看着这帮人费尽心机去对付一个女歌手，去对付她的歌迷，我只觉得替他们累，替他们难受。那种感觉，就像看一群猴子抬自己抬不起来的杠铃，龇牙咧嘴，丑态百出，还以为自己在展示神力。

更荒诞的剧本发生在演出之后。

Read more

This December, we’re teaming up with GiveDirectly to send cash to 800 impoverished families in the Bikara region of Rwanda. Studies show that direct cash transfers have a multiplier effect of 2.5x in local economies and reduce infant mortality rates by 48%. Your donation is also tax-deductible in the United States. The link to give is here, and the deadline for donations is midnight on December 31st. Please consider donating if you can!

is a researcher focused on semiconductors, AI, China, and Taiwan. He holds a Master’s degree in Regional Studies — East Asia from Harvard and was recently a summer fellow at the Centre for the Governance of AI (GovAI).

High-bandwidth memory, or HBM, remains the key bottleneck for China to catch up in manufacturing advanced AI chips. As Moore’s Law has more or less held steady, logic nodes have continuously progressed.

However, the rate of memory chip progression has been slow compared to logic chips. Thus, AI operations are often “memory constrained,” meaning that compute is sitting idle waiting for the memory chip to feed it data on which to perform operations. HBM was created to address this “memory wall” by stacking multiple memory chips on top of each other to boost memory bandwidth. As AI chips continue to get better, HBM remains a critical component for scaling. Simply put, if you care about the AI race and AI chips, then you must care about HBM.

Although China’s memory champion CXMT has been closing the HBM gap, the three memory giants of SK Hynix, Samsung, and Micron continue to be more than two generations ahead of CXMT’s HBM2. Assuming export controls hold steady, China’s HBM advances will continue to be stymied by a lack of advanced equipment.

For perspective, achieving the industry’s current HBM3E and HBM4 would be a tremendous achievement for China. As of November 2025, the most advanced AI chips in use use HBM3E. H100s, B100s, and other leading GPUs tap into HBM3E for memory, while Nvidia’s upcoming Rubin GPUs will use HBM4. If CXMT can achieve HBM4 quickly, then they will be able to crack a key part of making advanced GPUs. However, even if they are able to make HBM4 several years down the line, competitive AI chips will likely have meteored beyond contemporary standards to handle workloads unimaginable today.

Ray Wang’s piece earlier this year in ChinaTalk mapping CXMT alongside other memory giants helps policymakers keep an eye on China in the rearview mirror. But past HBM2, when will CXMT hit a wall? Given the current state of export controls and Chinese technological development, what node of HBM can China be expected to reach?

The Three Ingredients: DRAM, Base Die, and Packaging

Making HBM is a difficult endeavor, and the product’s performance ultimately comes down to three factors: the DRAM dies that compose the HBM, the base die that routes the signals coming in and out of the memory stack, and the packaging that binds the DRAM dies together.

Different bottlenecks exist within each of these three HBM components that will hinder CXMT’s progress at different HBM generations. Each merits its own discussion.

DRAM

The memory industry uses a different terminology to mark node sizes compared to the logic industry. Instead of referring to a node by nanometer, the DRAM industry has begun to use letters for its advanced nodes. They started first with 1x, then 1y, and then 1z; afterward, they moved to the Greek alphabet, with 1α after 1z, and then 1β, and then 1γ. (Samsung and SK Hynix use the English 1a, 1b, and 1c instead, but this article uses Micron’s terminology.) Just to demonstrate the gap between each generation, between Micron’s 1β and 1γ nodes, the product speeds increased by 15% while reducing power usage by 20%.

As of 2025, CXMT is three generations behind the leading memory manufacturers, making the 1z node while the big three are shipping 1γ. With the 1z node, however, CXMT can produce DRAM for HBM up until HBM3.

But what must CXMT do to achieve beyond the 1z node? To get to 1α and beyond, CXMT must shrink DRAM cells even further, which requires advanced tools in lithography, etching, and deposition.

Lithography

Two of the most difficult steps in DRAM manufacturing are forming the bitline contact (BLC) and storage node contact (SNC). The BLC is the physical connection between periphery transistors that decide what memory needs to be fetched to amplify their signals and the capacitors that actually hold the memory.

As shown above, patterning and etching the BLC must thread the needle so as to contact the source/drain of the array transistors rather than the buried wordline (BWL) shown in teal.

The case is similar for the SNC, the physical connection between the bitline and capacitor. As shown below, the SNC must be etched through layers of different materials to again connect with the source/drain of the array transistors, instead of the BWL.

As DRAM nodes progress, the pattern density and critical dimensions of these processes get stricter, and greater precision is required. Eventually, EUV lithography is needed for these processes.

However, Micron has used techniques like self-aligned quadruple patterning (SAQP) to continue to use DUV up until its 1β node. Chinese manufacturer SMIC has used similar techniques to stretch DUV use for advanced nodes in the past, like its 7 nm Huawei chip. CXMT is likely even better at utilizing SAQP given the memory industry’s lengthier history with the process. Even for 1γ, Micron only uses EUV for one layer of the process, likely either the BLC or SNC step.

Thus, CXMT can likely also stretch its DUV use until 1β. After that, considering Micron has attempted to delay EUV use until the last possible moment, 1γ and beyond will become extremely difficult without access to the export-controlled EUV equipment. Without EUV, advanced nodes will either be impossible to make or of terrible yield; according to some estimates, using EUV, while more expensive, saves about 3-5% yield for advanced nodes while decreasing process steps by 20-30%. Without EUV, CXMT’s progress in DRAM will likely be stalled at the 1γ node, meaning HBM4E and beyond will be difficult for China to achieve from the DRAM standpoint alone.

Etching

For etching, the picture looks more favorable for CXMT. Advanced etching is required for the steps above, as well as for creating capacitor holes. These holes, which hold the memory charges, have small critical dimensions, high pattern density, and are very deep. Etching narrow yet deep holes like this can lead to a variety of defects, shown below, and thus require advanced tools with high aspect ratios (ratio of height to diameter). Aspect ratios reached 40:1 in the 1x era, with estimates for advanced nodes closer to 60:1.

The U.S. has imposed export controls on advanced etching equipment, including anisotropic etchers (the ones needed for capacitor etch), though China has been able to domestically produce equipment defying the controlled parameters.

For etching through silicon nitride for the capacitors, BLC, and SNC, Chinese products include Naura’s Accura NZ and Accura LX, as well as AMEC’s Primo nanova. Technical specifications about Chinese products are not widely available, though the Primo nanova is specifically advertised for the 1x node and beyond. Although this means the product probably cannot be stretched to cutting-edge nodes, Naura’s tools may work well enough.

Regardless, the existing Chinese offerings demonstrate that China is not too far behind on equipment for capacitor etch. These tools are susceptible to having exaggerated capabilities or scaling issues with manufacturing, but, especially compared to lithography, they’re not so far behind. China holds 10% of the global dry etch market and is self-reliant for about 15% of its advanced etching needs. The country’s rapid growth in the industry also demonstrates that etching obstacles may not be so solid. In short, China’s HBM progress will probably not be meaningfully hindered by DRAM etching bottlenecks.

Beyond etching, advanced deposition tools are required for DRAM manufacturing, but the story is very similar to etching: China can already produce the tools required, so it will likely not be a bottleneck. China is self-sufficient for 5-10% of its deposition needs and is also rapidly accelerating its indigenization efforts.

Through-Silicon Vias (TSVs)

Another step in DRAM manufacturing for HBM is the formation of through-silicon vias (TSVs), diagrammed below. This front-end-of-the-line process forms the vertical connections that allow stacked DRAM dies to communicate and function together. Without TSVs, the concept of HBM and of nearly all advanced packaging would be impossible.

For making TSVs, the most important process again is etching. TSVs require precise etching through DRAM dies to later deposit the material that serves as the vias connecting all the wafers together. The U.S. has imposed export controls on etching equipment specifically for TSV formation (EC 3B001.c.4), but again, China’s domestic manufacturers have been able to defy these parameters.

TSV critical dimensions currently range from 3-5 µm with depths of less than 100 µm. As nodes progress, DRAM dies are getting thinner, and both the depth and CD will decrease. Currently, China already offers equipment to satisfy these TSV requirements. AMEC’s TSV300E advertises a TSV CD of down to 1 µm and can achieve depths of several hundred microns. Naura’s PSE V300, though not publishing its specs, likely achieves a similar performance. Chinese product specs may be exaggerated or with lower throughput, but empirically, TSVs do not seem to pose an issue for CXMT given its capacity rivals other leading memory makers.

Having already achieved likely self-sufficient capabilities in TSV formation, CXMT will not be bottlenecked from this step in HBM manufacturing.

High-κ Metal Gate (HKMG)

Another process difficult in DRAM manufacturing is implementing the high-κ metal gate (HKMG). As shrinking DRAM cells for performance gains becomes increasingly difficult, HKMG has served as another means to increase device speeds.

As shown below, periphery transistors on a DRAM die are normally advanced by shrinking distances between the source and drain while also thinning the gate insulator. However, when insulator thinness reaches its limit, leakage issues emerge, and HKMG is used to solve them.

HKMG replaces traditional gate materials in periphery transistors to accelerate electron flow and prevent power leakage. Partially due to implementing HKMG, SK Hynix was able to achieve a 33% boost in speed with a 21% decrease in power usage.

The HKMG process has been adopted by memory makers since, and CXMT is now beginning its adoption process too; however, some reporting indicates that CXMT is struggling with its HKMG implementation, leading to reduced yield and slower manufacturing ramp-up. Other memory makers have adopted HKMG in their process flows around the 1z node, where CXMT is stuck now, so the company must hurdle the HKMG barrier to keep pace.

Incorporating HKMG in DRAM processes is difficult, partially because of the simultaneous processing of the periphery and array on a single wafer. The thermal budget of the array, or how much heat the structures are able to withstand, is relatively low; this means that the standard HKMG processes for logic nodes cannot be so replicable for DRAM. Although CXMT is currently struggling with HKMG, this doesn’t seem like an insurmountable issue. The bottleneck seems to be the more amorphous challenges of experimenting and perfecting process flows rather than a concrete wall of equipment inaccessibility. The equipment required for HKMG generally relates to the deposition tools in which China seems more or less self-sufficient.

Because of the lack of “hard” barriers like lack of access to tools, HKMG adoption will likely not be a serious hindrance to China’s HBM advances.

Base Die

The HBM DRAM dies sit on top of the base die. Among other functions, the base die routes signals coming in and out (I/O) of the memory stack. Ultimately, regardless of how strong the memory dies are, the power of the base die determines the upper limit of memory bandwidth for HBM.

As HBM nodes have progressed, the number of pins on the base die has increased, along with the data transfer speed of those pins. As a result, memory makers have used more advanced DRAM nodes to function for the base die to satisfy the requirement. Around the HBM4 generation, though, memory makers are compelled to use more expensive logic nodes to handle the workload. As such, memory makers are now partnering with TSMC to manufacture their base nodes for advanced generations.

The advanced logic nodes used for base dies will pose a problem for CXMT in its HBM advancement. Without EUV lithography, SMIC has been struggling to advance beyond 7 nm without abysmal yield.

For HBM4, CXMT can retrace Micron’s steps and continue to use a 1β DRAM die for base die functions. However, this decision would have significant drawbacks. Not all HBM4 are created equal, and by using a memory-process base die, Micron has emerged with HBM4 worse than SK Hynix and Samsung. While Micron’s product meets the JEDEC minimum of 8 Gbps per pin and goes to 9 Gbps, SK Hynix and Samsung have been able to reach 10 Gbps per pin and beyond via logic node base dies. Micron claims that they have begun sampling HBM4 with 11 Gbps, but Irrational Analysis explains why this is probably misleading.

Regardless, Micron has conceded that memory nodes are not best suited for the base die after HBM4 and has partnered with TSMC to produce the base die for HBM4E on an advanced logic node. For CXMT, this likely means that using 1β DRAM dies for HBM4 will result in a subpar product, and that HBM4E will be difficult to make without SMIC making breakthroughs in logic nodes.

However, lower cost HBM4 and 4E may be possible for CXMT. Although memory makers are producing their most advanced base dies for HBM4 at 5 nm and below, they are also offering alternatives with cheaper 12 nm base dies. 12 nm base dies can get the job done, but the products with more advanced logic offer smaller interconnect pitches for memory performance and lower power consumption. These make the 5 nm base dies attractive for AI workloads desired by customers like Nvidia.

Although CXMT could theoretically partner with TSMC for its base dies, as they would likely not fall under export control restrictions, my conversations with experts suggest that TSMC may not accept such orders given geopolitical tensions. Essentially, without access to advanced logic nodes for the base die, CXMT will likely struggle to make competitive HBM4 and HBM4E. They will likely be able to make HBM4 with non-leading-edge 12 nm base dies. Perhaps they will even be able to secure orders from TSMC for advanced nodes, but the amount of question marks here makes CXMT’s success uncertain.

Packaging

Packaging is how the entire HBM stack comes together, and one element in particular is relevant. The “glue” that binds DRAM dies to each other, or bonding, is critically important. Stacking so many dies together creates thermal issues that bonding plays an important role in addressing; further, more efficient bonding with minimal gaps between dies is important to enable further stacking. As HBM has evolved from stacking only four dies to now up to sixteen, efficient bonding has been a key enabler.

Die Bonding

A possible struggle for CXMT will be succeeding in die bonding, but not because of export controls. Currently, export controls do not restrict the sale of bonding equipment used for HBM.

The two primary methods for die bonding in HBM are thermocompression bonding with non-conductive film (TC-NCF), used by Samsung and Micron, and mass reflow-molded underfill (MR-MUF), used by SK Hynix. SK Hynix adopted MR-MUF early on since HBM2E, and because of the decision, SK Hynix has been consistently lauded as creating superior HBM.

MR-MUF involves heating and connecting all the stacked dies at once, rather than one at a time like in TC-NCF. The real magic potion for MR-MUF, though, is the epoxy molding compound (EMC) used to fill the gap between dies.

MR-MUF has both better throughput and thermal dissipation than TCB. This is important both to scale production of HBM, but also to manage its heat requirements. By using MR-MUF, SK Hynix is able to stack more dies with fewer usage problems. HBM failures are the number one cause of AI chip failures, so MR-MUF to manage heat grants a real competitive edge.

Following SK Hynix’s footsteps, CXMT is reportedly adopting MR-MUF for its HBM3 and beyond; however, adoption is not like flicking a switch. To reap the benefits of MR-MUF, CXMT must solve several issues. First, MR-MUF is inferior to TC-NCF in managing die warpage. As DRAM dies become even thinner, CXMT will take time resolving this issue, just as SK Hynix has. SK Hynix solved this issue with a process it calls “advanced MR-MUF,” which adds a step of temporary bonding to the process — a step which CXMT may imitate.

Secondly, material acquisition may pose a problem. Competition, not export controls, may bar CXMT from acquiring the EMC for MR-MUF. SK Hynix has an exclusive deal with the Japanese materials company NAMICS for providing its EMC. SK Hynix’s material has been co-developed over years with NAMICS, and the material must be suited for each company’s process flow. Some Chinese sources suggest that CXMT’s EMC supplier is the domestic company Huahai Chengke (华海诚科), but this is still unconfirmed. Even if CXMT uses a domestic supplier, it will likely take years to work together to achieve a high yield.

Because of the extra steps from DRAM making to die bonding via MR-MUF, CXMT’s yield for its HBM3 in 2026 will likely take time to ramp up. Some experts claim that CXMT’s HBM3 yield likely won’t break 40% until the latter half of 2026, partially because of the MR-MUF adoption process.

In the end, though, CXMT’s early bet on MR-MUF will likely turn out to be a good idea in the long term, if not the short term. The advantages of the process are clear, and the bonding process only seems to be a short-term stumbling block. Though not a strict bottleneck, adopting MR-MUF will likely cause CXMT to slow production of HBM3 and beyond, but will not serve as a bottleneck for advanced generations.

Unanswered Questions

It is difficult to gauge CXMT’s capabilities or breakthroughs with 100% certainty. Unlike Chinese model developers, China’s chip manufacturers like to play their cards close to their chest. Because of the sensitive nature of their work, which is relevant for national security goals, or perhaps just because of the nature of the industry, CXMT rarely makes public statements. Perhaps this will change if CXMT undergoes its IPO as planned in 2026.

As such, certain details about China’s memory ecosystem are unanswerable without insider information. Some specific questions are listed below, and ChinaTalk invites anyone with color to reach out with answers or leads:

1. DRAM Node Sizes

   1. What are the critical dimensions of the latest DRAM nodes and their aspect ratios?

   2. What are the critical dimensions for TSVs in the latest HBM generations? How many TSVs are now included on a single DRAM die?

2. Chinese Equipment Ecosystem

   1. How good are AMEC and Naura’s etching equipment for mass production? How good is China’s deposition equipment in practice? How true are the advertised specs?

3. CXMT Struggles

   1. What part of HKMG adoption is CXMT struggling with?

   2. Who is CXMT’s EMC provider for MR-MUF?

If anyone has answers to any of these questions, or has information related to prior analysis, please respond to this email or reach out to jordan@chinatalk.media!

Conclusion

Overall, CXMT is progressing at a steady pace for making HBM, but this trend is likely not to hold forever. For each step of the HBM process — DRAM, base die, and packaging — different bottlenecks will appear to stall CXMT’s progress or compel them to make sub-par HBM. First, the lack of advanced logic for base dies will likely lead CXMT to make lagging-edge HBM4. Even if CXMT utilized a memory node for its base die for HBM4, this would result in an estimated 10% decrease in memory bandwidth. After HBM4, both the base die constraint and the lack of EUV for DRAM manufacturing will cause trouble.

Summary of Conclusions:

But CXMT should not be written off. The industry chose HBM as the best option for memory in AI chips because it was the path of least resistance. With export controls, that may not be true for CXMT and China. Other alternatives for alleviating the memory bottleneck have been discussed, including using hybrid bonding, high-bandwidth flash (HBF), a unified cache manager (UCM), compute in memory (CIM), ferroelectric RAM (FeRAM), and magnetic RAM (MRAM). All of these options have their own problems and are nowhere near adoption, but they present opportunities for China to move off the beaten path and achieve memory self-sufficiency in its own way. If any U.S. administration reverses export controls, though, China will be able to more quickly follow the path for HBM development and catch up in the AI chip race.

For now, though, with HBM remaining the preeminent option, CXMT will have its work cut out for itself.

, the new interim Executive Director of CSET who substacks at , makes her ChinaTalk debut. Present since the founding of CSET, Helen has had a front-row seat to the drama shaping today’s AI world — including a stint on OpenAI’s board.

Today our conversation covers…

• What it means to run CSET in 2025, and how to keep think tank work rigorous and relevant in the age of AI,

• The “good-faith” vs “dark arts” actors shaping Washington’s AI policy debate,

• What her recent trip to China revealed about how Beijing is thinking (and not thinking) about AI,

• Why AI progress might stay “jagged,” and what that means for AI policy,

• Plus: why Jordan can’t fall in love with AI.

Listen now on your favorite podcast app.

The 2026 Tarbell Fellowship is now open. You could come work with us at ChinaTalk! Apply here.

Don’t just take it from me. Take it from our current Tarbell Fellow, , on his experience so far:

“Tarbell placed me at ChinaTalk for a year, fully funded! It’s been a dream setup to report seriously on China, tech, and AI. The fellowship’s training covers both journalism and the fundamentals of AI, which makes it one of the best on-ramps for people who didn’t come up through traditional reporting or AI pathways.

I always thought about tech journalism but assumed I missed my chance after college. Tarbell gave me another shot. ChinaTalk gives me the freedom to chase questions I’m genuinely curious about in the China–AI space, paired with a team that constantly reads each other’s work, shares articles, and brainstorms ideas. You’ll be producing impactful work for a large audience, but you’ll also be learning every day.

At ChinaTalk, I spend my time digging into the semiconductor supply chain, Chinese AI models, U.S.–China relations, and whatever else I get excited by. If that sounds like your idea of fun, apply!”

Think Tanks in the Age of AI

Jordan Schneider: As the new interim Executive Director of CSET, are you excited to rip up everything they’ve created and remake it in the image of Helen Toner? What is your vision for the future of CSET?

Helen Toner: If there’s one thing that I have learned from the many friends and colleagues who’ve rotated in and out of government, it’s that your day-one mission should be reorganization. Step in, tear everything up, and change the org structure.

No, I’m kidding. It’s exciting and an honor to be in this position. After Jason and Dewey, I’m stepping into big shoes. I’ve been at CSET since its founding in 2019, so it’s exciting to shepherd the organization into a new phase.

CSET’s success is built on a foundation of excellent work, and I want to continue that. The core of our mission is to produce intellectually independent research that is driven by evidence and data.” Our data science team is unique in the think tank world — their data powers our analysis. On every project, we make sure our analysis is rigorous and driven by the best evidence we can find. We care that our work is technically informed.

One of our founding goals at CSET was to show a different way for think tanks to operate, and ideally, inspire others to follow us. I think we’ve been really successful there. You now see RAND with a huge emerging tech and national security effort, and CSIS doing more translations and data visualizations — things that were core to the CSET model and are now much more common in Washington.

That’s great, because it proves our model works. Of course, it also means we have competition, so we have to show what makes CSET unique and where we provide particular value. Our deep expertise on China is a perfect example. We have a whole range of China specialists woven throughout our team, covering everything from language to specific subject matter. I’m excited to lean into that and to keep evolving. Emerging tech never stands still, so we have to keep figuring out where we can add the most value.

Jordan Schneider: I agree that CSET has raised the bar for discourse in Washington — it’s why I gave CSET ChinaTalk’s only “Think Tank of the Year” award back in 2022. It’s been heartwarming to see your standard of using real evidence on thorny topics like chip controls, immigration policy, or the PLA’s use of AI resonate so strongly in the broader debate in Washington.

But at the same time, we’re seeing a paradox. Since 2019, it feels like facts matter less than ever. Arguments get reduced to tweet-shouting matches, and remarkably, those shouting matches are now becoming central to the actual policy debate on AI. What’s your take on these two trends happening in parallel? What’s the synthesis?

Helen Toner: I think there are multiple layers here. You have the headlines in the New York Times or the Wall Street Journal, but there is also work happening beneath the surface. The U.S. government has millions of employees, and the subject matter experts doing the work are interested in details and evidence. There’s a steady demand from them for the kind of support we provide, and they are very responsive to facts.

Another example is the discourse around recent AI legislation. Take California, for example. Last year, the discourse around the SB 1047 bill was awful. Then this year, they convened a governor’s panel, published a report, adopted its recommendations, and passed a less controversial bill. It’s a crazy turnaround. We saw something similar with the EU Code of Practice — it looked like it was going to fall apart, but then it came together. I don’t want to sound too pollyannaish, there’s a lot to be concerned about. But it’s important to remember that sensible work is still getting done.

Jordan Schneider: I started ChinaTalk in 2017, and CSET started in 2019. Back then, the intersection of U.S.-China relations, emerging technology, and national security was not a front-page topic.

Helen Toner: When we said we wanted to have a whole organization focused on emerging tech and national security, and people were like, “A whole organization? Like, four people?”

Jordan Schneider: It’s been a wild adjustment for this space to go from an idea funders would laugh at to something presidents tweet about all the time. But that shift has also brought in layers of bad faith. Back when this community was smaller, there weren’t many people playing dirty.

I think CSET has its heart in the right place and is doing earnest, yeoman’s work. But there are snakes in the grass everywhere now. There’s so much money riding on this research, and that wasn’t true a few years ago. I admire your pollyannaishness — I think it’s good for your mental health. But is the most effective option to put out good research and facts? Or are “dark arts” needed to have that research shape policy?

Helen Toner: I don’t think the only options are “put a white paper on your website” or “go full political dark arts.” There’s a lot of space in between. From the beginning, we’ve done more than publish research — we actively seek out the relevant policymakers, brief them, and work with their teams on legislation. Now, we’re also thinking about how the internet has changed what that means for us. Should we be doing videos? I’m not sure, but we should at least consider it.

Another big shift, which I know you follow, is the trend toward individual brands over institutional ones. Some of our people are eager to give that a go, while others — especially those from the intelligence community — are like, “Oh God, shoot me before you make me go on Twitter.” We’re exploring that space — finding ways to keep doing good-faith, fact-based work while operating effectively in today’s ecosystem.

Jordan Schneider: I worry that an organization where good-faith, facts-focused people are comfortable is fundamentally different from one with “dark arts” specialists. The cultures and incentives don’t mix.

Helen Toner: Will there be a ChinaTalk “Dark Arts Think Tank Award”? Who would that go to?

Jordan Schneider: Wow, I don’t know. I can’t give any names here — this is for public consumption. But I agree that there will always be an audience for grounded data, and someone needs to provide the facts.

Helen Toner: When we talk about “the Facts,” it’s not about some ideas being more virtuous than others. But if you want to accomplish something and care about results, then you need to know what the world looks like.

We worked closely with the Biden administration when they were considering outbound investment controls — asking them, “How will you implement these controls? Do you have the necessary information to do it effectively?” This isn’t about taking a holier-than-thou position. It’s about the reality that if you don’t know what’s going on, you’re going to try things that backfire — and most people want to avoid that.

Jordan Schneider: A better framing might be that it’s better to have data in the discussion.

It’s remarkable what a single researcher can do in recent years with an “individual brand”. It’s wild to think that CSET was around before we could ask ChatGPT what The PLA Daily 解放军报 says. I can now code data visualizations in two hours, which I used to assume would require a CSET-level team and budget. How do you think these new tools change what a solo researcher or a small team can accomplish? Does this change how CSET operates?

Helen Toner: We’re looking at it from the opposite side — what unique things can our larger team do that an individual still can’t? Our data team is tackling a huge data science problem called “entity resolution”. That’s figuring out that “Google London” and “DeepMind” are both “Google” in a massive dataset of text. It’s a huge, messy problem, and using language models in a carefully designed and validated pipeline, we blew past previous results.

We also analyzed ~3,000 AI contracts the PLA is buying and used language models to parse that data. As a larger team, we can do things an individual researcher can’t. We can validate our results and test different models — like when to use an expensive, frontier model versus a lighter one that’s faster and can handle high volumes. We’re doing tons of experimentation there, and the team is coming up with some really cool stuff.

Subscribe now

Jordan Schneider: CSET was early on important AI topics, but has remained ideologically neutral — you were writing about semiconductor export controls in 2020, but have not published an “AI will arrive in 2027” style analysis. Is now the time? Are the odds of those radical changes high enough that you need to start spending your team’s time and budget exploring them?

Helen Toner: I have a unique perspective because I have my feet in two worlds. I’m from the AI safety community, which is in that mindset, but most of my team at CSET is not “AGI-pilled.” We’ve done a lot of work on scaling and red-teaming, but not the “OMG AGI” work. We are currently hiring someone to work on frontier AI issues, and I’m hoping to increase our work in that space.

Jordan Schneider: What are you excited for this person to do?

Helen Toner: I’m excited about the “Frontier AI” framing. I’m glad RAND is now researching AGI, but the concept of AGI is messy and contested — it’s not clear what, if anything, is there. There has been a giant gulf between the AI systems we have — those we can touch and test — and hypothetical concerns about future AGI. But in two years, that gulf has gotten smaller. Now we can look at current systems and extrapolate future ones — which makes this topic amenable to CSET’s evidence-based methods.

I’m psyched for this research. It won’t be “CSET predicts AGI in 2027,” but it’s important to consider the possibility of AGI or superintelligence on timescales soon enough to matter for policy. Watch this space.

The Jagged Frontier

Jordan Schneider: You’ve expressed the view that AI progress could stay jagged. Can you elaborate on that?

Helen Toner: The idea comes from Prof. Ethan Mollick, and possibly also Andrej Karpathy. I highly recommend following Mollick’s AI work on Substack, LinkedIn, or Twitter. His idea was a “jagged frontier” — that AI is good at some tasks and surprisingly bad at others.

I recently gave a talk on this, arguing we should take seriously the idea that AI’s progress might remain jagged. Right now, most people fall into one of two camps — either they think AI is all hype and a “nothing burger” — or they’re in the “AGI by 2027” camp. That group believes powerful AI will become a drop-in remote worker or an automated AI engineer. Both are non-jagged views of the future. The question of what persistent jaggedness would look like is underexplored.

Jordan Schneider: The “jagged frontier” idea is more nuanced than mainstream discourse on AI — the Twitter brain, swinging wildly between “it’s over” and “we’re so back.” Why do you think people resist the possibility of uneven AI development — that the next model won’t solve everything? Why does the jagged idea struggle to gain traction, even though it is our current reality?

Helen Toner: Most people agree today’s AI is jagged, but they believe the future will be different. I think that’s because we use humans as a reference point — we believe that what’s difficult for us must be universally difficult, instead of seeing it as a product of our own evolution. Since the 1950s, we’ve debated — are we recreating the human mind, or building useful machines?

We’re currently far down the “build useful machines” path, but the idea of recreating the human mind is built into the field. I think this is why people expect AI to be more human-like than it is.

Jordan Schneider: There’s money involved now — the AI hype is backed by enormous financial incentives.

Helen Toner: Jaggedness doesn’t only refer to the troughs where AI struggles — there are high peaks as well. In the next 5-10 years, I expect us to exploit the heck out of those peaks. But the way we do so must account for the troughs.

Jordan Schneider: Is jagged AI more tractable for policy research? Is CSET’s work more relevant in that scenario?

Helen Toner: If jaggedness persists, fast takeoff scenarios are less likely — scenarios like an automated AI researcher that makes ten years of progress in six months. That would be a hard world for policy to operate in — there isn’t time for the government to form a commission, write a nice report, and debate it in the next legislative session. Jaggedness leads to slower AI progression, which gives us time to reflect, experiment, and adapt.

I’m not certain jaggedness will persist, but the idea is underrated in the AI community. At the same time, we should consider the possibility of non-jagged, rapid AI progress. It could still happen, although it’s not my best guess.

Jordan Schneider: There’s a resource allocation problem in AI policy research. Should we focus on a tangible, near-term jagged frontier — like AI’s impact on cybersecurity — or on the sci-fi futures of self-improving AI? People are drawn to speculative, sci-fi scenarios — a cybersecurity paper won’t go viral like “AGI by 2027” did. But there is value in working on a more probable future.

Helen Toner: There is a lot of low-hanging fruit in research on jagged development, and a lot of possible futures. What will AI be good at? What tasks will it struggle with? What does that mean for adoption and integration?

A jagged frontier means we are unlikely to fully automate complex jobs or goals. Instead, we will get powerful AI advisors and a “centaur” model of human-AI teaming, which you mentioned in the AI girlfriends podcast. Future human-AI collaboration scenarios are underexplored because predictions of super-powerful AI assume everything will be automated. They focus on abstract problems like alignment, not the messy, practical details of human-machine teaming that a jagged world would demand.

Jordan Schneider: After writing a paper on AI honeypot espionage, I decided to do some experimenting. Over the past few days, I’ve tried to fall in love with an AI, and it’s not lovable in the slightest.

When it comes to personal comfort and consolation, AI jaggedness is very apparent. There has been a lot of recent reporting about people who’ve developed close, intimate relationships with AI, but it’s not doing it for me. What should I make of that, Helen?

Helen Toner: Have you tried the Grok anime goth girl? You need to find the right one for you.

Jordan Schneider: It was bad — really repulsive. Even if I’m not the target audience for these AIs, if they were smart, they would have figured me out after 10 minutes of conversation — the way TikTok figured me out after 45 seconds of swiping. These models cannot do that — that’s an important detail.

The lack of personalized learning is a huge hurdle for AI in the workplace. Instead of learning from user input, models are trained and dropped into organizations, leaving people to figure them out. If the future of this technology depends on personalization that fits like a glove — professionally and personally — then we need to solve this.

Helen Toner: There’s a long way to go. We held a workshop in July about automating AI R&D and the potential for an “intelligence explosion” takeoff. We need to question underlying assumptions — what does progress look like? What are the gaps? How soon can we fill them? We’ll examine this in an upcoming CSET paper.

Jordan Schneider: There’s tension in our view of AI’s capabilities. It’s easy to overlook its limitations in work you do not do yourself, but in your own work, you can feel the jaggedness firsthand. You have an intuitive sense of where AI is exceptional and where it’s uneven.

Ironically, AI engineers are the most optimistic about AI’s capabilities — maybe a little high on their own supply. But the proof is in their paychecks — companies are hiring them in droves because AI cannot do their jobs.

Helen Toner: There are many sources of jaggedness.

A key source of AI’s jaggedness is the context window — how easy is it to input the organizational or practical context of a task? Some professions, like software engineering or marketing, are easily digestible for an AI because you can copy-paste the relevant code or creative brief. But most jobs can’t be reduced to a text file — their context is messy and organizational. We haven’t fully grasped how this single limitation will shape what AI can do and how quickly it can do it.

AI Debates in China

Jordan Schneider: Helen, you were in China recently. How was that trip?

Helen Toner: It was great to be back in China. In 2018, I was in Beijing for 9 months, studying Mandarin and learning about China’s AI ecosystem. But between my green card, the pandemic, and having kids, it had been ages since I was there. I went for a quick five-day trip to Shanghai for the World AI Conference, which was gigantic. You know what Chinese conferences are like — the huge stage, the flashing lights. Robots were walking around everywhere, something you couldn’t get away with in the U.S. Kids were petting little quadruped robots that were roaming the floor. It was a good time.

Jordan Schneider: Were you recognized?

Helen Toner: No, definitely not. Not that anyone told me.

Jordan Schneider: What’s your sense of the U.S.-China AI dialogue and opportunities for discourse or cooperation?

Helen Toner: People in the AI safety community often ask why there isn’t a U.S.-China dialogue on avoiding a race to superintelligence. The answer is that there is no agreement on what the problem is, or what the U.S. and China’s interests are. At a Chatham House discussion I recently attended, the Chinese organizers were divided on whether to focus only on superintelligence or broader development questions as well. Within their team, there was no consensus on the core issues. These conversations are a good start, but we still have a long way to go.

Jordan Schneider: A core AI policy question is how the U.S. and Chinese ecosystems will relate to each other. What are the other key questions that will define the field for years to come?

Helen Toner: On the national security side, the U.S.-China dynamic is a big one, covering both competition and the potential cooperation on AI. Military integration is another huge question. The focus is shifting from developing advanced AI to how it changes a military’s operational concepts and the way it fights. This is an adoption challenge.

There are also serious risks around cyber and biosecurity, but we might get lucky, and the threats are manageable. I’m personally more concerned about cyber, but I know well-informed people with access to classified information who are deeply worried about the bio risks.

Outside of national security, we’ll see more community-level issues, particularly around data centers. A narrative about their water use is gaining traction, and while the data may not show an unusual amount of consumption, the community perception is strong enough to create backlash. There are also social questions. We do not have a framework for dealing with AI companions, especially for children, and the impact of AI on labor and jobs is not going away.

AI Parenting Advice

Jordan Schneider: Do you have any AI parenting takes, Helen?

Helen Toner: I have a three-year-old and a one-year-old, so thankfully, we’re not there yet. But I worry the “engineer-brained” approach to parenting reduces child-rearing to a set of tasks. The idea that if an AI can entertain or teach a child “better” than a human, then it’s a net win, misses the point. The relationship between a parent or teacher and a child is a huge part of what it means to grow up and learn. AI should be a tool to enhance connection, not replace it. If an AI generates a story, read it to your child, but do not be too utilitarian. What are your thoughts?

Jordan Schneider: Abstracting love is a high bar for AI.

Kids are wired by billions of years of evolution to trust a warm, sweaty mammal. An AI can certainly teach them physics or math better than I can, and outsourcing that is one thing. But the biological need for connection is another. Primate studies show the same thing — the monkeys want to be held. Trying to engineer that need away is playing with fire. Maybe a robot will get there in 20 years, but you’re running hard against evolution. No offense to anyone using Midjourney for children’s books — I have that tab open right now.

Helen Toner: I think there are good ways to do it.

Jordan Schneider: Absolutely. But the sci-fi future where kids don’t need loving parents for connection or as models of how to relate to other humans seems a long way off.

Helen Toner: There is a This American Life story that sticks with me, about a single dad and his daughter. He was a physicist, and she would ask him astronomy questions like, “Why do stars...?” or “Where did the Earth come from?”. Kids love to ask “why” questions. He found answering them stressful, so one day, he asked her to write down all of her questions. He locked himself in his office and wrote up a gigantic set of answers for her. The interviewer on the show asked the girl what she thought, and she said, “I wanted to hang out with my dad.” It’s so tragic. Don’t do that with AI.

On Calling Timeout

Jordan Schneider: My theory is that CSET only exists because of Jason Matheny. The national security risks of China’s rapid AI growth were completely off the radar for these funders. It took an exceptional person they trusted, like Jason, to convince them to build a community around this idea.

Before CSET, there was no tech team with deep China expertise. I spent years trying to make the case that competition with China mattered, that AI was more than one small piece of a larger puzzle, but people were unconvinced — that idea was ’too spicy’ or too far out.

There was a brief moment during the 1st Trump administration when it became a mainstream concern. Many corporate blogs, including that famous OpenAI document, were suddenly about beating China. But that moment has passed, and it feels like the issue is becoming less relevant again.

Helen Toner: It’s an interesting time for China+AI policy. When I started in this space around 2017, people in AI would ask, “Why talk about an AI race with China?” and then give AI-specific reasons why it wasn’t a race. I had to explain that they were missing the bigger picture. The U.S. national security apparatus was orienting towards strategic competition with China. For them, AI was only one small manifestation of that competition, and the AI community’s arguments were seen as irrelevant noise.

Jordan Schneider: I remember people telling me, “Oh, but if we say this, will it accelerate the race?” Bro, come on.

Helen Toner: Is strategic competition with China still the main goal of the U.S. national security apparatus? People outside the tech world are not sure — the current U.S. policy toward China is unclear. That’s disconcerting in some ways, but it also creates potentially productive space.

Jordan Schneider: In the 1st Trump administration, U.S.-China competition was a central pillar — Jake Sullivan wanted “as large a lead as possible.” But in Trump 2.0, focus on China has waned, and now we’re mobilizing against Venezuela while ignoring Chinese boats in the Philippines.

For years, I’ve thought that the U.S.-China AI race was inevitable. In many ways, it has already happened — we now have bifurcated ecosystems for AI chips, models, and hyperscalers. These dynamics seem resilient to the day-to-day whims of policymakers. How durable is this rivalry? If the American president is not focused on this issue, do the competitive dynamics of the last decade have enough momentum to continue on their own?

Helen Toner: I don’t know how resilient the rivalry will be. The competition was never about AI — competition with China was the organizing principle of the U.S. national security apparatus, and AI was one part of that. The U.S. AI sector now uses that narrative as justification for everything from faster data center permits to avoiding AI regulation. If those arguments lose force, I’m not sure what will happen.

My own prediction has always been that China’s internal demographic and economic challenges would eventually cool the rivalry, though I thought it would take longer, maybe till the 2030s. With a president who is hard to predict and an increasingly isolationist MAGA base, and a new focus on the Western hemisphere, disengagement with China could be stickier.

Share

Jordan Schneider: The TikTok story is a…

Helen Toner: Hilarious episode in the sitcom that we live in.

Jordan Schneider: Exactly! This was a bill Congress passed almost unanimously, and then the president decided he was not concerned with an issue Congress had a bipartisan consensus on — that is an interesting detail. I’m not sure how illustrative that is.

Helen Toner: Another source of tension in the Trump coalition right now is between the “tech right” and MAGA. They have disagreements about whether to charge ahead with AI — whether AI is the best thing since sliced bread, or the devil, or the Antichrist. There is a lot of division, but both sides are less concerned about competition with China. The tech right wants to sell to China, and the MAGA world would prefer to slow the rate of development.

Jordan Schneider: Corporate self-interest could be a reinforcing driver. U.S. firms do not want to compete with Chinese companies on their home turf. Once Chinese EVs start taking market share, or Huawei chips threaten Nvidia, the game changes. The question will become, is access to China’s market worth giving up our own? The likely answer is no. U.S. companies will demand the same protected home base that Chinese firms have used to their advantage, which only accelerates the competition.

Helen Toner: Wait, they can operate here? I thought it was a one-way street.

Jordan Schneider: Bill Bishop often invokes a Xi Jinping quote that essentially says, “Our goal is to become more self-reliant at home and make the world more dependent on us.” This mindset was in the rare earths saga, where China’s escalation was a self-inflicted wound. It showed a willingness to compete in a way that alienates American elites. You can admire their ambition, but the U.S. will not accept Chinese competitors dominating key verticals — especially in the tech sector that underpins the U.S. stock market.

Helen Toner: We’ll see. I do not know how we can ban Chinese open-source models, which I think is one of the biggest threats to U.S. market share. Using open-source Chinese models presumably displaces API market share for OpenAI, Anthropic, or Google.

Jordan Schneider: They are not un-bannable — stopping individual downloads of Chinese software is a fool’s errand, but that’s not the real game. The real game is preventing billion-dollar companies from being built on Chinese open-source models, and the government has plenty of ways to do that. They can block Chinese models from government contracts, tie it up in FCC compliance issues, or make it a mandatory risk disclosure. If the U.S. government really puts its back into it, it can find a way.

Helen Toner: The government procurement restriction is a good point. Public company disclosures — that’s interesting. I agree, these policies can make it harder.

Jordan Schneider: Or change the incentives. Switching tangents, can you pitch some of the best recent work from CSET — what do you admire and plan to build on?

Helen Toner: One of our most exciting new papers analyzes 2,800 PLA AI contracts. The initial piece focuses on who is buying, and the key finding is that while the largest contracts go to state-owned enterprises, the bulk are awarded to “non-traditional” private companies and universities. More research is coming on what they’re buying.

Our work on DoD AI integration has also been impactful. Interestingly, our research has been valuable to government officials because it is public. Internal reports are often classified and hard to share, so a URL they can circulate is a game-changer. Our paper “Building the Tech Coalition,” which analyzes their use of Project Maven and the internal talent required, is a great example of this.

Number three is our work on AI and biorisks. The debate has been narrowly focused on controlling AI models, so our “Toolkit for Managing Biorisks from AI” broadens the conversation by outlining a full range of policy options, which has been helpful for policymakers.

Jordan Schneider: Let’s do two more, oldies but goodies.

Helen Toner: For oldies but goodies, I’d point to our outbound investment work, where we asked the Biden administration, “If you want to control outgoing investment, do you know how to do that? What data do you have, and what data would you need?” That implementation was a classic example of our work.

Our explainers have been surprisingly impactful. We published one about the differences between generative AI, large language models, and foundational models. A government agency was trying to decide which terminology to use in an influential policy document, and told us the explainer directly influenced their policy. Straightforward research like that has a good track record.

Jordan Schneider: Would you like to recommend some mood music to end the episode?

Helen Toner: The great China and AI scholar, Matt Sheehan, told me instrumental playlists are the best way to focus, so I’ve been listening to a lot of instrumental music. There’s a great James Brown instrumental album. Why not some instrumental James Brown?

Mood Music: