Takeaways from the Trump-Xi Summit: The Economic Security Angles

This transcript is from a CSIS event hosted on September 28, 2026. Watch the full video below.

Scott Kennedy: Good morning, or good afternoon, or good evening wherever you are. Welcome to this CSIS program, “Takeaways from the Trump-Xi Summit: The Economic Security Angles.” I’m Scott Kennedy, trustee chair in Chinese business and economics here at CSIS, and I’m glad to invite you to this discussion about the Trump-Xi summit.

I think as everybody knows unless you are living under a rock, that last week Xi Jinping, China’s leader, visited the United States and Washington for a visit with the president, and the two sides engaged in discussions about a whole variety of issues. This was their second meeting of the year, the first coming in May in Beijing, and it could be the second of four with potential encore presentations in Shenzhen in November on the sidelines of the APEC leaders meeting and in December in Miami on the sidelines of the G-20.

A central part of their conversations and negotiations revolved around economic security issues, and you’ve probably had the chance to hear from a lot of China experts already telling you what this means for U.S.-China relations. Today we’re going to hear from experts on different elements of economic security, technology, clean tech, dual-use technologies, supply chains, advanced technologies like semiconductors to see how this summit, what it means to them and how they interpret the implications of the summit for U.S. foreign policy and the global economy.

Before I introduce the panelists and turn things over to them, just a brief summary of some of the takeaways, just a factoid; not trying to steal any thunder from my excellent colleagues who are going to be speaking with you in just a minute.

But people may have noticed very early this morning the U.S. Trade Representative Office issued details about the Board of Trade and how it is operating, its structure, and clarifying the commitments on both sides to identify $30 billion of goods on each side for lower tariffs. They said that they are considering lowering tariffs, and when they said reduced they didn’t specify the specific number. But we’re talking about 14.4 percent of total U.S. trade in 2025, if that’s what they’re working with.

In addition, there were commitments based on the fact sheets from both sides related to sales of agricultural goods, fish, timber, coal. Didn’t hear anything specifically about aircraft.

They mentioned standing up the Board of Investment, though no specific details on that.

There was an announcement for the creation of a dialogue on superintelligence. That’s the American fact sheet. The Chinese statement still uses the word “artificial” intelligence.

There was discussion on Iran mentioned by both sides. There was discussion of Taiwan, but not mentioned in either of the fact sheets. I think people may have also noticed U.S. Ambassador Perdue’s comments on American national TV yesterday about some elements of the Taiwan discussions.

Maybe perhaps most importantly, the Chinese committed to give two pandas, Ping Ping and Fu Shuang, to the United States, and they have already arrived in Atlanta. That’s probably the quickest deliverable implemented from the summit.

And then, finally, the two sides mentioned that they are – have intentions of attending the APEC and G-20 meetings later this year.

So that’s the basic list of deliverables related to economic security, drawing no judgments good/bad/otherwise about them.

To talk with me about these issues, I’m delighted to have four of my colleagues with us.

Kate Koren is deputy director of the Economics Program and Scholl chair in international business at CSIS. And before coming to CSIS, she was with the U.S. Department of Commerce Bureau of Industry and Security, part of a long, distinguished career in the U.S. government.

Lauryn Williams is the deputy director and senior fellow in the Strategic Technologies Program here at CSIS. And until January of 2025, she was chief of staff to the assistant secretary of defense for industrial base policy within the Office of the Undersecretary of Defense for Acquisition and Sustainment.

Sujai Shivakumar directs the Renewing American Innovation Program here at CSIS. Previously, he directed the Innovation Policy Forum at the National Academy of Sciences, Engineering, and Medicine.

Ilaria Mazzocco is deputy director and senior fellow within the trustee chair here at CSIS. And she is one of the world’s leaders in understanding Chinese clean-tech policy and industry trends, and what it means for the rest of the world.

Thank you all for being with us here this morning on the program. And what I’d like to do is just first turn to each of you for a few minutes of your recap and analysis about the summit and its various economic security elements. So why don’t we first go to Kate, and then Lauryn, Sujai, and Ilaria in that order. Kate, thanks for joining us, and over to you.

Kate Koren: Thanks so much, Scott. Thanks for putting this together today. And thanks, everyone, for listening.

So, for me, you know, the one of the topics I follow most closely, particularly in relation to U.S.-China relations, is dual-use exports, export controls, and export controls on advanced technology. So last year at the Busan Summit, the détente was agreed to, and that was set to expire in just a few weeks here. And looking forward, when we were kind of looking at what was going to come out of this summit, I think myself and many others expected an extension of that détente, particularly related to export-control issues. Those had been quite contentious and being used sort of interchangeably with different tariff rates and other economic changes as a – leverage points in a tit for tat, going back and forth between each other.

And in Busan there was an agreement to, you know, basically put a – put a hold on that, stop that, and that included – you know, we didn’t get full text on that, but what that amounted to in practice was not just the U.S. pulling back, delaying its affiliates’ role – which I’ll go into in a little more detail in a second – but also appeared to really be a hold on any export-control policy changes directly affecting China. Those included ICTS import controls directly affecting China. Those also appeared to be paused from this détente and any real significant sanctions actions and entity listing actions from the Department of Commerce against Chinese entities. All of those went into kind of a freeze, you know, in that past year, the affiliates role being the one with the actual public, you know, direct text acknowledgment.

So the affiliates rule was a change to the way that BIS operates its entity list. Those are a list of entities that U.S. exporters are not allowed to export to without a license, which is generally denied. The affiliates rule would have changed it so that it’s not just the specifically named entities on the list, but any entities that are 50 percent or more owned by those entities. And the practical effect of that would be to add quite likely thousands of additional Chinese entities into that sphere. So this was definitely a point of contention in U.S.-China relations in this issue, and one of the tangible outputs of Busan last year was the U.S. agreeing to delay the implementation of the entity list rule for a year. That was set to snap back in November.

So it’s not surprising to me to see that that has been extended. It’s a little surprising to see that it’s only been extended till January. So the full détente is extended till January. We don’t have the official regulatory text pushing back the affiliates rule implementation, but we have heard it’s coming. There’s some off-the-record U.S. official remarks that this is included and it’s coming. The interesting thing is that it’s such a short amount of time.

So what we’re hearing is that that is because the U.S. feels that China isn’t living up to its end of the bargain on the exports of rare earths, that they’re not allowing enough of those in the amount of – you know, the amount of ease of export in return, that they’re not living up to their end of the deal, and want to continue to put the pressure on China to meet its end of the bargain there before doing a further extension of the affiliates role. So that’s keeping things a little interesting.

I know the length of the trade truce in general has been met with some consternation, I think, by, you know, some in industry because it doesn’t leave a lot of – it doesn’t give as much kind of comfort and stability as some would like. But it’s interesting that these are still – these are still kind of live issues. You know, my expectation is that in these later meetings throughout the year there will – agreements will be reached and it will be – continue to be extended, but you know, you never know. It’s still – it’s still live.

Let’s see. The other notable thing, really, to me from this particular summit is just the real absence of, you know, those kind of – the topics of, you know, chip sales, advanced semiconductor sales. You know, previous summits led to, you know, the allowing H200 advanced GPU exports. That really seems to have all been either off the table or at least, you know, very much not for public discourse on this one. And there’s been a few encouraging statements from USTR and others about specifically separating out those national security topics from commercial deals. The tariff reductions, USTR says, are being worked through the Board of Trade and are explicitly only about commercial goods. And the topics related to advanced tech, anything touching AI, all those sort of things are being done in a national security forum that is – that is separate, which I find encouraging, think that’s a good – a better approach to not kind of commingle those issues. But, also, we don’t know where those stand right now.

So back to you, Scott.

Dr. Kennedy: Kate, that’s super good. And just to foot-stomp what you just mentioned, this effort to differentiate economic security/advanced tech from other parts of the commercial relationship, if you look at the list of goods in the 30 for 30 that USTR pushed out recently it’s basically consumer goods and raw materials and things like that, so- but not rare earths. We’re talking, you know, agricultural products and other things like that. Thank you so much. Really appreciate that.

Lauryn, how – what are – what are your takeaways and analysis from looking at the summit?

Lauryn Williams: Thanks so much, Scott. It’s good to be here this morning and to have the opportunity to debrief on these issues with great colleagues. So great to follow on from the context that that Kate set.

And for me, so I, of course, sit in our Strategic Technologies Program, which is looking at the range of emerging technologies and their economic security and kind of national security implications. So it will surprise no one that I was most focused on the AI- – or superintelligence-, from the U.S. readout terminology – -related deliverables or conversations, as well as the context – and I’ll actually spend maybe more of my time talking about that – that framed the environment for Trump and President Xi to meet this past week. So the AI dialogue I’ll talk about, as well as the – such as there was conversation, and in the readout we see information around the broader context that the Iran conflict provided for these meetings. So those are the two issues that most stuck out for me.

So, on the AI front, if, you know, you’re here in the United States, you know that the context that we are living in when it comes to the technology and its fast-moving, kind of rapid-moving capability advancement is leading to a lot of national discourse in a way that we, you know, haven’t seen in a long time around an emerging-technology issue. So that means, of course, that even just in the last couple of weeks there have been calls by leading U.S. frontier developers, tech executives here in the United States, to use Dario Amodei and Anthropic’s terminologies, “pace the frontier;” to really think about making sure that companies, as well as the government, at least in his call and in his written remarks, are working together to understand and to implement guardrails, and even make sure that the government has a role to play in governance for this technology. So we’ve got these calls. We have, of course lots of conversation across the United States around datacenters, which are not unrelated, and broader kind of artificial intelligence anxieties across the United States. Pair that with the latest examples that are kind of trickling out daily from OpenAI’s in-testing models to Anthropic’s models to your Google’s or hyperscaler, and frontier model companies are kind of providing new evidence and new indications of their most advanced models either accessing the internet or demonstrating capabilities that are beyond what might have been expected in this point in time. So I’ll just lay all of that out there to note that for that reason there was a big spotlight and a big interest for me personally and I know for many experts on what might come out of this dialogue from an AI perspective.

So what did we see? And I think we have to step back a couple of days from the summit itself and look at conversations, a dialogue to put it that way, that Secretary Bessent and the Department of Commerce, as well as his counterpart – his Chinese counterpart, He Lifeng, had earlier in the week in New York. And then we even had remarks from President Trump in his U.N. General Assembly remarks that kind of provide context, and then I’ll talk specifically about the deliverable language itself from the summit.

So what we heard and what we saw about the discussions that Secretary Bessent in particular had were that he and his counterpart discussed a proposal for a, quote, “notification mechanism” between the U.S. and China, particularly on national security-related incidents. So this was maybe one of the more kind of concrete proposals that was put on the table last week, which I think really does speak to the heightened – the heightened focus on this issue here in the United States and then, of course, in China, with the U.S. and China being the top – the top technology developers.

What I think is notable about that announcement such as we’ve seen details on it, it seems to be kind of a direct line of communication between these two leaders. And for those of us who have spent time in – time in government, you know that kind of only the most significant issues will reach – find their way to a principal, and principals also tend to not have as much time as their staff to focus on these issues. But it does demonstrate, on one hand, the seriousness with which both sides are taking these issues, but it does also raise questions about implementation.

Secondly, to kind of move forward in time, President Trump, of course, in his U.N. General Assembly remarks expressed the tone that has been consistent from him over the last several weeks, which is, one, his desire to change the terminology around artificial intelligence and to connote it superintelligence; but then, number two, to really – to keep the technology kind of sprinting ahead – from the United States’ perspective, not to limit innovation. And so his specific remarks included language around really not wanting to hamper innovation, and so that’s been his consistent line.

Heading into the summit itself, moving forward in time, what we saw explicitly in the readouts from both governments was not what I might have expected, because there had been discussion out of the May summit that the sides might hold a – and I’m maybe putting words into the statements’ mouths – a fulsome dialogue around AI-related issues in conjunction with the current summit. What we saw was an agreement – and both sides articulate this – to hold a future summit on AI safety-related issues, ostensibly in November of this year around a next meeting point. And so that’s kind of the substance of what we learned around the two sides’ discussions on AI issues.

As Kate already mentioned, there – in previous summits, of course, there’s been a lot of discussion around the sale or not of advanced semiconductor chips to China, which have generally been under very tight export controls. But we saw a thawing or kind of a change of heart from the Trump administration in expressing its willingness to sell NVIDIA’s most advanced, or H200, chips to China a couple of months ago, but there hasn’t seemed to be much progress on operationalizing that. So that being the other key issue that we didn’t see discussed.

And then, finally, just to touch on Iran, which is of course a really important context, it’s one that that I’m interested in from a geopolitical perspective but also given revelations that we’ve seen in the media in recent months around reporting that, you know, China is providing intelligence to Iran that may or may not be useful in targeting or kind of enhancing its capabilities there. And then, of course, we know that China receives a significant amount of oil imports from Iran and has remained a partner of Iran’s even in the current conflict environment.

And so the – finally, the language around discussions such as they were around Iran were essentially that, from the U.S. side, the leaders would discuss ongoing global concerns, to include Russia, North Korea, and Iran. They agreed that Iran can’t have a nuclear weapon, and then further, specifically to the context of the conflict, that no country or institution can be allowed to impose tolls on international waterways. But I think the language that we see there I would extrapolate to say that it’s fairly muted, not least because of the broader desire for both sides and particularly for the United States to maintain kind of the fragile or tentative détente that we’re seeing on trade issues.

So, with that, I will stop and hand the floor back to you, Scott.

Dr. Kennedy: Thanks, Lauryn. Very good review of several of the issues related to AI, chips, and Iran.

I’ve got a variety of questions I want to come back to, but I first want to go to Sujai to get his take on the summit and all things high tech, which is what his program works on here at CSIS. Sujai?

Sujai Shivakumar: Thanks, Scott.

So I think the question was, how did – you know, how did the summit turn out with regards to what’s on my radar screen? What I’m looking at when I look at the radar is, you know, what’s the larger picture in terms of innovation policy. And you know, largely things have remained intact. I don’t think – I haven’t seen anything, you know, major coming out of this particular summit. You know, there’s no formal agreement or progress made towards easing high-end U.S. technology controls. Even though we had people like, you know, NVIDIA CEO Jensen Huang and AMD CEO Lisa Su at the White House state dinner, you know, export controls didn’t serve as really a primary sticking point or a major topic of debate during the direct talks.

You know, there are some issues about China easing their – you know, their restrictions on certain types of chips coming, you know, for some of the high-end – well, the not-so-high-end NVIDIA chips, which is, you know, NVIDIA had this interesting challenge that U.S. was not allowing and China was not receiving those chips. But otherwise, you know, it was not too much change as to my perspective. I think, you know, China is actually benefiting from a lot of the export control in terms of that inadvertently, from our perspective, stitching together their innovation system both in terms of their chipmakers and chip users. And also, I think the fact that we have our midterms coming up, I think the Chinese probably understood that they should just wait and see at this moment rather than make major commitments. So not too much change, from my perspective.

Dr. Kennedy: Sujai, that’s very helpful.

Ilaria, I didn’t hear the word “climate” mentioned too much during the summit, but obviously clean-tech issues must have been discussed and a variety – energy, et cetera. How did you evaluate the summit?

Ilaria Mazzocco: Yeah. Thank you, Scott.

I think I agree there was a lot that was unsaid, or rather that didn’t appear at least in the readouts. I think there had been tons of speculation both in May and again now in September that there might be some conversation around the automotive sector, right? So there was speculation that Ford might join the delegation in Beijing and that, you know, BYD was said to – the CEO of BYD was said to be planning to join the state dinner the other day. Instead, we got some very sort of weird reporting around several Chinese CEOs getting on a plane, coming to D.C., not being invited to the state dinner, sort of hanging around D.C., and you know, we can imagine they headed back home at this point with some disappointment.

I think that is – it’s not just related to clean tech. I think this relates to the broader challenge of Chinese investment in the United States. I think, despite the fact that at least the U.S. readout mentions the Board of Investment, we really have no indication that they’ve made significant progress in this area, which remains fairly controversial, I would say, in the United States at least. So I think in a way what’s remarkable is that in the readout and in also the documents that we’re now seeing related to the Board of Trade we don’t really see any of the technologies that everybody’s very concerned about, except superintelligence AI, where we’re really talking about a dialogue, right? There’s no sort of commercial or sort of more regulatory agreement around that. So I think, you know, everybody here in D.C. and certainly in Beijing is worrying a lot about all the more advanced technologies, right – biotech, robotics, electric vehicles, batteries, et cetera – and that’s just nowhere in this framework.

When it comes to energy, there’s a relatively broad statement on oil, which is more of a remark that the United States made or encouragement the United States made towards China. And then there’s, you know, the U.S. readout mentions coal, that China will increase its coal imports from the United States. That’s certainly something that China is capable of doing. I think there’s – you know, we’ll have to see how that plays out. Of course, there’s been sort of spotty implementation of some of these bilateral trade deals. But that’s about it, right? We don’t really see anything else.

So it’s true that, I think as my colleagues mentioned, this is now – I think we’re seeing this meeting as part of a broader set of meetings. So any sort of broader deal that might come out of this – and I think we should certainly be, you know, careful not to have too high expectations around what kind of deal might happen – probably won’t happen until December, right? So we’ll have to wait and see.

But I think, you know, this just points to the fact that a lot of structural challenges remain, right? I think we have sort of the stable moment where the two leaders can meet and have pleasant dinners, you can have sort of deals around non-sensitive areas. You may have maybe dialogues around more significant, you know, areas such as AI. But I think, you know, some of the big, core structural issues between the two countries, as Sujai said, really haven’t changed at all.

So I’ll stop there.

Dr. Kennedy: Super. That’s very, very helpful.

Thanks, all four of you, for your initial quick takes. I thought of some general questions I wanted to ask and shared them with you all before, but because you guys said some really helpful, thoughtful things, let’s follow up with some of the specifics.

And I’m going to follow up first on AI, because that was really significant that they apparently devoted a lot of time to it in New York. They said they had the first meeting, which was – as Lauryn said, was Bessent and He Lifeng themselves talking, the principals talking, and said that they would have another meeting by November.

I guess a couple things about this. Superintelligence, is that an invention of President Trump, or where does that come from?

And the second has to do with the notification systems and hotlines. Lauryn worked in – you worked in the Department of Defense – or is it Department of War? I’m not sure what we’re supposed to call it – where we’ve had a variety of hotlines with China and other countries over the years. And I’m curious, is that going to be, you know, secretaries, Cabinet members carrying cellphones around or looking for emails from their friends or going back and direction? How would a notification system work if we wanted it to work? Is there going to be a red phone that people are calling for this kind of thing? And what would count as something that would deserve notification?

Ms. Williams: No, thanks, Scott. I think those are – those are super-interesting questions, and some I can – I can shed light on from past experience and maybe others I’ll be speculating on because we don’t have a lot of detail, including on your – on your last question, of what might reach the threshold of being an incident that these two – the two countries talk about.

So, firstly, on the terminology “superintelligence,” I’m glad that the president did – I think it was in his UNGA speech and maybe before that – actually articulate what his rationale was or what his thinking was, because he might think – I might think it was one thing, but he did actually say that he prefers the term “superintelligence.” It sounds like something he came up with himself, I think, one, because super is a superlative that he prefers, but also because he thinks and kind of takes “artificial intelligence” to mean fake intelligence, kind of taking the literal translation of the word “artificial.” So that’s what he said in terms of why he’d like to use that terminology.

And I think it is interesting as well that kind of in comparing the two readouts from the two countries China, I think, while noting that – I think – I was going to say while noting that they would use that terminology, but now I’m not seeing that in the text, they did say that the two sides agreed on the establishment of a China-U.S. AI dialogue. So that’s the Chinese readout. The U.S. side says both countries agreed to refer to the term by the – refer to AI by the term that President Trump raised. So that’ll be interesting to see going forward what the official line is.

And then, on a notification system. Yeah, you ask a great question. This is an idea that, I think, for those of us who’ve kind of been in the defense and broader technology spaces, kind of harks back to maybe nuclear weapons, nuclear arms control. But it also, to me, when I was working in the Pentagon on space policy issues there was talk around kind of coordination of kind of satellites that get too close to each other, communicating between the two sides, in particular U.S. and China have the most satellites. So that was a context where we talked about these issues.

And what would be important I think for either side, for a project like this to actually work, would be to make sure that – you know, the details, talk about the two, essentially, secretaries of commerce being on either side of the phone, but you would really need staff. You’d need people whose job it is to monitor issues, to be aware of the broader environment, and to isolate issues that should rise to the level of a senior leader’s attention. So that really does hinge on having a bureaucracy on both sides that is designed around this to ostensibly kind of filter issues up to the top, that reaches that threshold.

And then, you know, to the important question that you ask, which issues or which incidents should rise to that level? I don’t think – I don’t think that’s been spelled out yet. But you can imagine, as we’ve, you know, seen details flow out about OpenAI’s models in particular in recent weeks, that have escaped their testing environments and reached the open internet, and hacked into other countries. And we’ve heard even about a case in which Australian networks were targeted, that – you know, in an incident in which a pre-released model is able to crack into a company that causes a concern – or, another network that causes a concern for kind of the U.S.-China bilateral relationship, that that might be something that filters up to these two leaders.

Dr. Kennedy: Thank you for that. That’s very helpful.

Does anybody else want to comment on this? Feel free to jump in, Kate or Sujai. Sujai, go ahead. 

Dr. Shivakumar: Sure. No, I think part of this, you know, rebranding it as superintelligence is a way of trying to calm passions down a little bit. I think, interestingly, both parties are now campaigning against, you know, certain ideas of AI. And we don’t want that to get too far ahead of ourselves because we will have climbed up the tree and down a branch, and this may be, you know, some way of kind of calming the waters. Also, in terms of our conversations with the Chinese, you know, telegraphing that we will, you know, somehow control the – you know, the – some of the fears and apprehensions of people who are worried about, you know, these models getting out of their sandbox, and so forth. So I think it is more of an atmospheric approach towards trying to, you know, calm people down about the fears about this issue. 

Obviously, you know, artificial intelligence, superintelligence is out there. And this is going to be the basis on which we are going to be competing, you know, going forward. So we don’t want the political, you know, impediments coming in the way of that progress. So I think having that – toning the conversation down, I think, is a useful thing at this point. 

Dr. Kennedy: Yeah. Kate, did you want to add to that? 

Ms. Koren: Yeah. So on the rebrand, it’s interesting that he particularly, you know, used “superintelligence.” And, you know, I think, you know, Lauryn’s exactly right that it’s – I think he came up with it, but it is an existing term in the industry from a 2014 book, that is a bit of a doomsday-type scenario book in some ways, a book by a philosopher, and defining superintelligence as, you know, a type of AI where it has become much, much smarter than any human. You know, cognitive abilities beyond what a human is, which is not where we are. There’s no indication that that’s what the president is saying, and using the term in that way. I think he’s, you know, using it in a – he’s sort of redefining it. 

But it is an existing term that, you know, AI researchers had and were using in a different way. And it’s interesting that he’s seems to be using it almost in an opposite way, in some in some ways, because what we’ve mostly been hearing from the president is a denial of the AI doomsday scenarios. So it stuck out to me, the recoining of that term. (Laughs.)

Dr. Kennedy: Well, we’ll have to see if this rebranding, how long it lasts, if this was just sort of a test effort to see how it goes. And if it sticks then maybe we’ll see more of it. But it’s clear that they just took some initial baby steps toward having a conversation and seeing what happens. And, you know, many aspects of the summit were baby steps. So a little more details in other – some places than others. But this is really sort of very, very incremental across the board. 

Let me –

Dr. Mazzocco: Scott, can I? 

Dr. Kennedy Yeah, Ilaria, go ahead. 

Dr. Mazzocco: I was just going to say one very brief thing. What struck me about this AI dialogue or superintelligence dialogue is that it’s actually the only sort of global governance area where now the U.S. and China seem to have any sort of agreement to at least want to cooperate in in any sort of practical way, right? There were some statements over Iran or some other statements, but I think this seems to be the only area. And I thought that was interesting. 

I went back to see the 2015 state – the readout from the state visit under Obama. And it was entirely global governance related, right? There was nothing related to bilateral trade. There was, you know, some cybersecurity-related issues, which I think were slightly more bilateral, but it was very much framed as a global governance issue. So I think it’s just sort of interesting where relations have really shifted over the past decade. And I think it’s going to be interesting to see if the two countries can actually set up anything more significant on AI, because it would sort of be indicative that there can be at least some cooperation on global governance at this point. 

Dr. Kennedy: Yeah. Well, certainly in 2015 one of the big takeaways at that summit that President Obama hosted was new commitments that both sides made on climate and carbon. And as I alluded to a little while ago, those words were absent from the conversation. It doesn’t mean that the Chinese have stopped, right, with trying to address their economy’s dependence on fossil fuels and trying to become a leader in green tech, as you document on a daily basis for all of us. 

Let me widen out the aperture a lot. All of us in the EST at CSIS follow economic security issues broadly, and the administration’s approach. And I’d just be curious, instead of – instead of just evaluating the specifics, the sort of broadly the Trump administration and its way it thinks about economic security issues and how it’s doing. So on a scale of one to 10, if you had to give a number about how well they’re doing on economic security issues – not just with the summit over the past few days but the last year and a half – what’s the number that you all would assign? Let’s start with – we’ll go Sujai, Lauryn, Kate, and Ilaria.

Dr. Shivakumar: Sure. Well, at least from an innovation policy point of view, it’s not a very high score in many ways. You know, you’re signaling two things simultaneously. On innovation, a lot of emphasis being put on manufacturing. But at the same time, you know, the issues like research and development skills are now being hampered, in many ways. So, you know, and I think, Scott, you described this as putting one foot on the brake and one on the accelerator. So, you know, it’s a very inconsistent set of policies. And so the – I think the – you know any innovation system is a network of different types of policies, all, you know, to some various extents working together. You don’t see that here. And so that’s a cause of concern. 

Dr. Kennedy: All right. So, Sujai, I still have to ask you for a number. 

Dr. Shivakumar: Maybe – well, it depends on – it’ll be below five, certainly. (Laughs.) Maybe –

Dr. Kennedy: All right. So we’ll take less than five.

Lauryn.

Ms. Williams: Yeah. Well, maybe we’ll all converge around the same number average. I was going to say five, but I think I might trend down. But I actually – I like Ilaria’s point a little bit earlier about thinking about at least the most recent spate of bilateral meetings between President Trump and Xi in a broader context. Because I think I would ask you, Scott and Ilaria, if, you know, this has been, kind of, the most frequent – or how this compares in terms of frequency of China and U.S. presidents meeting, maybe even within the calendar year. It seems like it’s fairly frequent. 

So taking it in a broader context, yeah, I’d say maybe five or 4.5, because we’ve at least kind of seen the moving out of the kind of thorniest – or the extension of the thorniest trade issues, which I think would otherwise take up all of the oxygen in the room. And it leaves at least openings for discussions on, you know, AI in particular going forward. I would like to see a meeting, a substantive meeting, happen in the future, as opposed to commitments for meetings, as we’ve seen out of the last two summits. 

Dr. Kennedy: Kate

Ms. Koren: Yeah. I think I go right down the middle with a five as well. You know, not a lot of substance. Not a lot of real movement. But not a lot of movement in a bad direction either, right? Just sort of – like, we are just sort of settled in the middle. Yeah, it’s – I can’t really think of more to add on to what Sujai and Lauryn have already said so well here. There’s not a lot to dig into and to grade right now. 

Dr. Kennedy: OK.

Ilaria.

Dr. Mazzocco: I’d probably go with a four. And I’ll note that I went to grade school in Italy, where it is a one to 10 grade. And below a six, you have to do remedial exams, otherwise you don’t pass on to the next grade. I will note, I think it’s true that there hasn’t been a lot of change. And yet, that’s a decision, right? That buys time, but that also gives time. So I think that’s just something to keep in mind on both ends. And so that, I think, creates – you know, if it’s not done in a strategic way, I think that could be a challenge, from an economic security perspective, right? So I think that’s just worth sort of keeping in mind there. 

Dr. Kennedy: Well, perhaps also, you know, I don’t know about Italy, but in the United States oftentimes we grade on a curve. And so it’s about how you score relative to others. So we’re not going to do that complicated math and compare President Trump to multiple other presidents, et cetera, but I’ll take note that we’re around four to five is the consensus so far. Expect detailed essays and reports about this score in the next week from all of you, because I know you don’t have anything else to do. 

OK, let’s turn the angle now to allies. Let’s say you’re sitting in Brussels, Paris, London, Seoul, Canberra, Ottawa. How do you look at what happened last week? You’ve historically had very close relations with the United States. Collaborate on a lot of issues, economic, military, economic security. Those relationships have been tortured and difficult the last few years because of tariffs, and also efforts to try and press allies to do more with their own defense and take other steps. And the, you know, relationship and how Xi Jinping was treated looks a little bit different than how some other leaders who’ve visited the U.S. have been treated. 

Just as an example, we had Xi Jinping and his group in the Oval Office. No Zelensky-like moment. And that was because there were cameras, but no journalists there to ask questions, so there were a lot of happy smiles. I’m sure that that was negotiated in advance. But if so, if you’re looking at this from the perspective of allies, do you come away reassured that the relationship is being maintained and not re-escalating? Or are you concerned that the U.S. seems less interested in pressing some of the challenges that the world faces with China? What do you think allies are – the lessons that they’re drawing? Why don’t we start with Ilaria, and then Kate, Lauryn, and Sujai.

Dr. Mazzocco: Yeah. Thank you, Scott. I think, you know, there might be some elements that might reassure allies, such as probably, you know, the fact that the two countries are discussing AI safety, although some may feel excluded. I think the – but I think, you know, at a broader level, it’s probably going to be concerning, right, that there is still no, you know, long-term resolution on some of the issues that threaten everybody in terms of global stability, such as rare earths. And I think, you know, we can probably expect that to continue to be an issue for the foreseeable future. And at the same time, you know it’s interesting that the – you know, this this trade deal, if it goes through, will see a reduction in tariffs in certain products up to a certain amount, while a lot of allies will be feeling that they are still under significant pressure when it comes to tariffs with the United States, and are under a certain level of threat of increased tariffs over time. 

So I think that’s just going to be, I think, something that’s going to be noted among allies. I think, you know, this is also – just, you know, very recently you had conversations between Canada and the European Union. I think very symbolic overall, not necessarily things are going to change at a structural level. But I think, you know, that’s just sort of the general trend, that you might have allies thinking that they need to look elsewhere if they want more stable economic partnership. 

Dr. Kennedy: Kate.

Ms. Koren: Yeah. Very much agree with everything Ilaria just said. Just try to build on that a bit, if I can. You know, I think for allies and others just around the world, you know, stability in the U.S.-China relationship is good for everyone else, particularly as, you know, China’s retaliation, particularly in regard to rare earths, against U.S. policies wasn’t limited to the U.S. So there has to be, I think, some amount of, you know, relief to see, you know, continued stability in that relationship. But also some consternation in just the optics of the visit. You know, the somewhat stark comparison to the sort of effusive, you know, pomp, and also just the amount of how much President Trump expressed, you know, the closeness of his relationship and his respect for President Xi, compared to others who would have not expected to be treated less well than the Chinese president in the past. 

So I think it’s a mixed bag. And, you know, again, just echoing Ilaria on I think there’s probably relief that the U.S. and China are having, you know, the AI dialogue, but along with some concentration that’s not moving faster. You know, we hear a lot from, you know, Brussels in particular of really wanting to be moving faster on you AI regulation, AI global governance, but being hamstrung by the realities of where the technology sits. You know, the understanding of, you know, you can – if they regulate now, there’s nothing particularly for – there may be nothing for them to regulate. 

Dr. Kennedy: Sure. I think the EU wants the U.S. and others to move faster on AI. The U.S. would like them to move faster on tariffs. It’s always the – that seems where we are. And maybe though in the next few weeks we’re going to see some movement in that, based on what’s going on in Europe right now. So we’ll see. So maybe that will be realized. Lauryn.

Ms. Williams: Sure. So definitely not much more to add on top of what Ilaria and Kate have already said. Would just foot-stomp the point that I would imagine that our allies and partners are glad at least to see a commitment to dialogue by the two, you know, leading AI technology developing nations right now. So I think that’s a point of positivity. But then, more broadly, I think what is maybe most stark about the – you know, both readouts, whichever way you slice them, is the very, I’ll characterize it, kind of inward-looking tone that’s struck in each one of them. Obviously, this is a bilateral summit, and so to an extent it makes sense that we would be mostly focused on bilateral issues. But there is not much, if any, kind of discussion of broader issues aside from the two wars ongoing in the Middle East and in Europe. So I can imagine that that is a read that our – in particular, our European allies and partners are drawing from it as well. 

Dr. Kennedy: Let me turn now to the final topic of our conversation. This has been really illuminating and helpful. 

In November, it looks like President Trump will make his way to Shenzhen. First time a U.S. president has ever made two trips to China. Looks like in December Xi Jinping will get on the plane and come back for the G-20 in Doral. Again, the first time a Chinese president has ever been on U.S. soil two times in the same calendar year. I think you also mentioned the extension of the ceasefire was pushed out to January 10th. We have a variety of dates that we can circle on the calendar. 

What are you all going to be looking for over the next few months about these dates on the calendar, and other elements that are on the economic security agenda for the United States? What do you expect to see? What would you like to see? We’re heading to the holiday season as well and so it’s time to ask for some gifts of where the U.S. and China could make progress, either bilaterally or beyond – because APEC and G-20 are multilateral gatherings. So there may be a chance to raise some of these issues to a broader level. 

Let’s start with Sujai, then Ilaria, Lauryn, and then we’ll give Kate the final word. 

Dr. Shivakumar: Thanks, Scott. So I think – you know, I think the thing that I’m watching most for is the U.S. midterm elections. You know, I think the APEC summit occurs right after the U.S. midterms. So if the election yields, you know, domestic political headwinds for the Trump administration, you know, President Xi may leverage U.S. political vulnerability to demand, you know, steeper technology or tariff rollbacks. We have to wait and see. I just wanted to also just have a chance just to weigh in on the previous question in terms of, you know, our allies. 

Our allies, like us, are trying to de-risk supply chains. But what’s been happening in the U.S., particularly over the past two years and more, is sort of a move of industrial policy, R&D policies, much more towards the states. A lot of the manufacturing that’s going on is very state-focused. And many states are looking to focus their bets on particular technologies. You know, Illinois and Colorado on quantum, you know, North Carolina on biotech, et cetera. And so what our allies are increasingly doing is to actually reach out to the state level. So, for example, New York is reaching out in a big way through IBM and other partners to build out the – you know, the semiconductor ecosystem in Japan. So, you know, I think there are different – it’s moving a little away from Washington. Our focus has been too focused on bilateral in terms of what’s happening across state capitals, but there’s a lot more under the surface. And I think we need to be watching those as well. 

Dr. Kennedy: That’s helpful. I hinted at thinking about going up to the multilateral, and you said go down to the local level, which I think is super helpful. I really appreciate that.

Let’s go to Ilaria. 

Dr. Mazzocco: So I’m going to be watching the board of investment. I don’t have very high expectations, but I think this would be quite significant if we did see any movement on it. So I’ll certainly be tracking to see if there’s any progress on that. I think beyond that, I think any sign that the two countries can sort of talk about things that go beyond just the sort of, I sell you this, you buy that, you know, and talk more about the global – I don’t necessarily want to say global governance – but bigger issues that transcend just the sort of immediate questions, I think that would be welcome. So certainly the AI dialogue, but if anything else emerges I think that’s going to be very important. 

Dr. Kennedy: Super.

Lauryn. 

Ms. Williams: Sure. So, of course, I will be eagerly waiting to hear developments in November when and if there is a bilateral summit on AI safety. But more importantly, given that, you know, we’re at the end of September now, November is not very far away. It’ll be important to have the – you know, the two commerce secretaries and counterpart meet. That demonstrates the importance of the issue. But what is really more important is the – you know, the staff work that goes on in between now and then to make sure there’s a robust agenda and make sure that there, you know, can actually be substantive technical conversations. And that happens at the – kind of the civil service level, but it could even happen at, you know, a track two or track 1.5 with non-governmental organizations engaging with Chinese counterparts. All to make sure that the two sides have a good understanding of each other and their technologies, which will lead to a better summit at the higher level. 

Dr. Kennedy: Kate.

Ms. Koren: Yeah. So I’m going to be watching with great interest to sort of see where the U.S. sort of viewpoint, really the president’s viewpoint, on AI safety, AI regulation, kind of shakes out between now and then. You know, there’s been a lot of public discourse, you know, between Dario Amodei’s post about the dangers and the need to pace the frontier and to slow down, one of the OpenAI researchers, you know, resigning with lots of doomsday warnings. So there’s been a loud – a loud voice on one side. And what we’ve seen so far is the president really rejecting and saying that those are, you know, overblown – those fears are overblown. 

I’m very curious to see how or if, you know, that thinking changes over the next few months, because I think that will be really key for what proceeds in terms of a U.S.-China agreement on any of these issues. If we tilt towards one or the other, if we are in the full – where the U.S. position is that these fears are overblown, unless safety dialogue, safety issues are needed to be addressed, I think that will have – it can’t help but have a large influence on how those meetings go. So very curious to see how that goes. It’s been interesting to watch sort of what seems to be sort of a coordinated effort from, you know, the U.S. leading AI companies to bring these safety issues, you know, more to the forefront, but it’s not – doesn’t seem to have quite captured the president’s view just yet. 

The other very sort of nitty-gritty thing I’m watching, date wise, is the affiliates rule that I was talking about earlier. You know, that’s set to snapback November 10th. We’ve heard it won’t, but there’s – it technically should require a published regulation, although we may see that addressed just with an announcement instead. You know, as they did with the rescinding the AI diffusion rule. But I’m curious to see if there’s an agreement to extend that further that would allow that date to be pushed back further than just a couple months at a time. 

Dr. Kennedy: All right. Let me take a second to try and answer my own question, because I’ve been totally unfair and un-reciprocal. I have asked you all hard questions. I haven’t been asked any hard questions, or even easy questions. So let me see if I could take a little bit of this, and also try and pull a few threads together. And I’m not going to say a lot that I think is different. 

So I also will be looking at the midterms. I think that will have an effect on – contextually on how the U.S. is thinking about the relationship, and what level of political space the administration has. And also then, if the Democrats win the House and/or Senate, what that might mean for the kind of issues that are pulled up to the national agenda, and what’s discussed nationally. I’m also looking for not what people are saying about AI, but what is happening in artificial intelligence and technology, and in companies and labs. 

It feels like a million years ago that people were talking about Mythos. Oh my God, that was like we were – that was the issue in April and May. And that’s, like, long gone. And last few weeks we were talking about Hugging Face. What’s going to be the thing that we’re talking about in a couple months about incidents, or the race in large language model capabilities? Because the shift really changed the conversation about what we wanted to have around artificial or – excuse me – superintelligence. So I’m looking just what happens in the world of technology and stuff, again, outside D.C. 

I’m also really looking for to see what happens in China. China’s economy, it’s slowed down dramatically. They have a giant credit crunch, debt problems. Their exports are booming while their economy is slowing down dramatically. The U.S. is prepping for midterms. The Chinese are prepping for the 21st Party Congress, Xi Jinping’s fourth term. Same time you have purges in the military and elsewhere. So that is affecting how confident or unconfident they are in pushing back. And I think all of that for me then gets to the point where Kate and all of you mentioned that you’re looking, it’s about the ceasefire. 

You know, only two months. But let’s say we get a few more months down the road. We’re in the holiday season. But people are starting to get nervous. Does either side feel confident about what the world would be like if we took away the guardrails of the ceasefire and change people’s expectations about the relationship? And, you know, is the U.S. ready to deal with a world in which rare earths would be restricted again heavily? Like, have we built up stockpiles, alternative sources, et cetera? I’m not sure. How would China feel if there were real significant restrictions on exports in lots of places, or the U.S. decided to really implement the sanctions regime that it just announced vis-à-vis Iran or related to Russia? 

Where are we? Will we be in a point where both sides have to grin and bear it again and extend? Or are we going to see some kind of snapback? And when you go around the halls of Washington, you hear people talk about the snapback option as, as soon as we’re ready that’s what we’re going to do. But I think we’ll end up being in a very challenging position if it gets to that. So as calm – as polite as things were the last few days, I think it’s going to be increasingly challenged to keep that tone as we get to later in the year. 

I want to thank everybody for watching today this discussion about the summit and economic security. But I mostly want to thank Sujai, Lauryn, Kate, and Ilaria for joining us and sharing your wisdom with everybody. These issues obviously are not going away. We’re going to be watching what happens between now and the next meeting in November, and the one in December. And we will be commenting, talking, and analyzing what’s happening. So wish everybody a good day, wherever you are. Thanks so much.

 (END.)