Did the U.S. Just Hand its AI Edge to China?

0:00Support for this show comes from Strawberry.me. Be honest. Are you happy with your job? Or are you stuck in one you've outgrown? Or never wanted in the first place? Sure, you can probably list the reasons for staying, but are they actually just excuses for not leaving? Let a career coach from strawberry.me help you get unstuck. Discover the benefits of having a dedicated career coach in your corner. Go to strawberry.me slash unstuck to claim a special offer. The world is buzzing with AI tools. But instead of making things easier, they've made things overwhelming. there's a better way. Enter Superhuman, the AI productivity suite that gives you superpowers so you can outsmart the work chaos.

0:50With Grammarly, Mail, and Coda working together, you get proactive help across your workflow. Experience AI that meets you right where you are. Unleash your superhuman potential today. Learn more at superhuman.com slash podcast. That's superhuman.com slash podcast.

1:13Avoiding your unfinished home projects because you're not sure where to start? Thumbtack knows homes, so you don't have to. Don't know the difference between matte paint finish and satin, or what that clunking sound from your dryer is? With Thumbtack, you don't have to be a home pro. You just have to hire one. You can hire top-rated pros, see price estimates, and read reviews all on the app. Download today. Today's number, 75. That is the percentage of top-grossing movies that are now sequels and reboots. That number is up from 25 % in 2000. More proof that Hollywood's greatest special effect is copy and paste.

2:08Welcome to Prof G Markets. I'm Ed Elson. It is December 10th. Let's check in on yesterday's market vitals. The major indices were muted ahead of the Federal Reserve's interest rate decision due today. Notably, the Russell 2000 hit an all-time high. Meanwhile, the yield on 10-year treasuries was stable. So was the dollar. Bitcoin rose. And finally, news broke in after-hours trading that SpaceX is pursuing an IPO in 2026. According to reports, the company is seeking to raise more than$30 billion. That would make it the biggest listing ever. Okay, what else is happening? In a sharp turn for US AI policy, the US is greenlighting advanced chip sales to China.

2:53President Trump has announced that NVIDIA can sell its H200 chips to China in exchange for a 25 % cut for the US government. He claims that President Xi Jinping, quote, responded positively to the proposal. Nvidia stock climbed around 2 % in response. However, after that announcement, the Financial Times reported that Beijing is already moving to limit access to those chips. By Tuesday morning, Nvidia gave back those gains and closed in the red. Okay, here to discuss what is going on here with these H200s and our AI chip policy, we are speaking with Chris McGuire, Senior Fellow for China and Emerging Technologies at the Council on Foreign Relations.

3:34Chris, thank you for joining us on ProfG Markets. Yeah, of course. Thanks for having me, Ed. So we've gotten this announcement from the president. NVIDIA can now sell H200 chips to China. But this is the next development in a story that seems to keep on going in multiple different directions. First, it was that we can't sell the chips. Now we can. Then we can't. Now we can. What is the timeline here? What has actually happened? And how has this news changed our policy on whether the U.S. can or cannot send chips to China? Yeah, that's a good question. I'll just walk through quickly the history here.

4:16So it was starting in October 2022, the United States first implemented policies to say the most advanced AI chips that were actually drawn in line by A100s, the level below this one, the previous generation. Anything more advanced than that cannot be sold to China. The policy at the time was to maintain as large of a lead as possible over China. And there are basically two prongs to that. One was maximally slowing China down in AI, and the other one was maximally investing at home. So that has been the policy. There's been various kind of ways to tighten it and close loopholes over the years.

4:52But the other prominent one is at the beginning of the Trump administration, NVIDIA consistently kind of made chips that just worked around. And the Trump administration in April of this year kind of closed all the loopholes, said NVIDIA can't be in the China market. But then quickly reversed that policy and allowed the H20, which was a much less advanced chip to be sold to China. Now they are taking a much more significant step to say that the H200 can be sold to China, which is the first chip that is over all of the original thresholds. This is the first time that we're really exporting AI chips that are far greater than the intent of the controls, and that really would provide a pretty substantial plus-up in China's compute capability.

5:38It's six times better than the H20. It's nine times over the export control limits. So even though it's the previous generation chip, it's not the Blackwell chip that's the most advanced. It's still a very, very capable chip, and it's by far the most capable chip ever sent to China. So that's kind of a timeline here of how all this has played out. Obviously, big questions about what does this mean in the future, but also kind of, you know, why does this matter and what are the implications that we can talk through? Yeah, I appreciate that timeline. Sounds like under Biden, we had restrictions, but looser restrictions.

6:13Then we tightened a lot. Then we slightly loosened. Now we're slightly loosening again. I guess my question is, why? Why are we now loosening a little bit more? What was the impetus for this? Yeah, I think actually all the changes before now were more on the margins. The big changes, I think, were October 2022 when the United States first implemented these controls and first said AI chips to China are going to be banned. And then the action yesterday where we said, actually, we are willing to export advanced AI chips to China. So those are like the real bookends. Everything else in the middle, I think, is more the kind of details of the story.

6:54The rationale of the control was because large numbers of advanced chips are necessary to make advanced AI models, and China cannot make advanced chips themselves. This is a significant advantage the United States has over China and AI. It's probably the only significant advantage the United States has over China and AI. I think on talent, algorithms, data, electricity generation applications, it's either a watch or China's ahead, but hardware is the area where we are very ahead. There actually has been pretty significant bipartisan support for this policy historically. But Jensen Huang has argued very vehemently, particularly over both administrations, but particularly effectively in this administration, to lessen those policies.

7:38A valid question as to why. I think from NVIDIA's perspective, they definitely want to have a more diverse customer base. I think three companies probably make up 50 % of their revenue. That poses an understandable strategic issue for anyone who leads a business. So they want to be able to sell to more providers, but that probably necessitates them selling to China. Up until now, the U.S. government has said that's not in the U.S. national interest. That might be in NVIDIA's interest, but we think that's going to materially advance China's AI capabilities, which poses national security threats.

8:14And therefore, we're not going to let it happen. I think that President Trump deciding this has upended that and is basically saying whatever is good for NVIDIA is good for America. Personally, I think NVIDIA is a fantastic company that produces amazing products and is doing great things. but I think that I probably wouldn't make as blank of the statement, personally. Yeah, I guess that's the part that is interesting, is that it seemed that this was all about national security. We can't do it. We can't sell our chips to China. It's too important. It's too dangerous. And now we're deciding, yes, we can do it.

8:46And I guess the thing that's confusing as a viewer, as an observer, from the administration is, why? Why now? Is it that Jensen is now buddies with the administration? That's certainly what it begins to look like when we see all these dinners and maybe we're being, I don't know, too blunt there. But that's certainly what it appears. I mean, what is it, do we think? Why is the administration deciding that this national security concern, which seemed like such a big deal, is no longer as big of a deal? I think it's a very good question. And I do think the answer to why this happened is it came down to personal relationships.

9:24And I think, you know, Jensen Huang has convinced the president that this is a good thing. I think there are many people in D.C. on Capitol Hill in the administration who are very skeptical of this, who do not believe that we should be sending China our most advanced technologies and have concerns about how China can use AI in the military and defense and espionage realms to undermine U.S. national security. has active plans to do so. So I think, look, there are other arguments that people make for exporting the chips. They'll tell you, well, if we don't export them, Huawei will fill the vacuum.

9:59I think that one I disagree with because actually if you look at the US, NVIDIA and Huawei's AI chip roadmaps, NVIDIA's chips are getting substantially better. They're already five times better. And in two years, they're going to get to about 20 times better. China's chip making capability is very constrained. It is the one thing that they can't make really well. Sophisticated AI chips, they're the most complicated thing to make in the world. So I don't think they'll fill the gap. The idea that we could get, you know, country, China addicted to U.S. chips. I also just don't think there's been any technology where China has ever accepted being reliant on a foreign power, particularly, you know, they have been adamant not to be reliant on the United States, in particular, on semiconductors.

10:43So I don't think there's any putting the toothpaste back to the tube on this. So look, there are arguments, but I personally don't think they carry as much water or board out by the history of the data. I think the reason is because personal relationships have pushed big policy decisions across the line. Yeah, it is very interesting just from a national security, what is best for America perspective, where on the one hand, it's, well, we don't want to be giving China chips to advance their AI efforts. And then on the other hand, well, do we like that we have this leverage? Do we like the fact that they are so dependent on us?

11:18And I think what is very interesting is this more recent report that we heard from the Financial Times, which is that Beijing, Xi Jinping, China is actually actively limiting how many of these H200 chips their own companies can buy. So in other they're saying we actually don't want to buy them we're going to try to restrict ourselves we're going to cut ourselves off from this what do you make of that report yeah that is very interesting and we've seen this play out uh multiple times over the last few months i mean when the united states initially allowed the h20 so the the kind of less good ai chip to go to china um china has blocked that and i think that was in my opinion very likely a negotiating plight i think They were trying to push for the more advanced chip, and that ploy worked.

12:09I do think there is an element here where China is trying to make sure that they protect their own market. They are trying to make AI chips. Huawei is making AI chips. They're not nearly as good. They're not making nearly as many, but they are trying to do it. And China will make sure those chips get sold. They're not going to let those chips get unsold and just stuck in warehouses. So they will manipulate their domestic market to say, hey, these these are the things that we need to use domestic chips for and make sure that the Huawei chips, you know, can continue to all get sold and be used. And that will be the case no matter how many chips they make.

12:44But yeah, the question of like whether or not they actually, I'm sure they do have concerns about like backdoors and security, but whether or not they would actually be willing to forego the very substantial plus ups in compute capability that sending H200s would represent. I'm a little more skeptical, but, you know, countries make silly decisions all the time. So I don't know. Christopher Guaya, Senior Fellow for China and Emerging Technologies at the Council on Foreign Relations. Chris, this is fascinating stuff. Thank you for joining us. Thanks very much for having me. After the break, Google's next product, Smart Glosses.

13:25If you're enjoying the show, give Prof G Markets a follow.

13:35Support for the show comes from Vanguard. To all the financial advisors listening, let's talk about bonds for a minute. Capturing value and fixed income is not easy. Bond markets can be massive, murky, and let's be real, a lot of firms throw a couple flashy funds you're wearing call in a day. But not Vanguard. Vanguard's been in the game a long time, and their scale gives them a serious edge. They're able to invest across all kinds of sectors, maturities, and geographies, which means they can spot and act on opportunities that others might miss. Some managers out there promise big returns, but those usually come with big risks, and that can mean a rollercoaster ride for investors.

14:12Vanguard takes a steadier approach. It's not always flashy, but it sets the standard for what dependable investing should look like. So if you're looking to give your clients consistent results year in and year out, go see the record for yourself at Vanguard.com slash audio. That's Vanguard.com slash audio. All investing is subject to risk. Vanguard Marketing Corporation Distributor. Support for the show comes from Public.com. You're thoughtful about where your money goes. You've got your core holdings, some high conviction picks, maybe even a few strategic option plays on the side. The point is, you've engaged with your investments and public gets that.

14:52That's why they built an investing platform for those who take it seriously. On public, you can put together a multi-asset portfolio for the long haul. Stocks, bonds, options, it's all there. Plus, an industry-leading 3.8 % APY high-yield cash account. Switch to the platform built for those who take investing seriously. Go to public.com slash prop G and earn an uncapped 1 % bonus when you transfer your portfolio. That's public.com slash prop G. Paid for by public investing. All investing involves the risk of loss, including loss of principal. Brokered services for U.S.-listed registered securities, options and bonds, and a self-directed account are offered by Public Investing, Inc., member FINRA, and SIPC.

15:31Complete disclosure available at public.com slash disclosures.

15:55T-Mobile.

16:04mobile network in the U.S. based on analysis by Google Speedtest Intelligence Data 1H 2025.

16:11We're back with ProfG Markets. Google is taking another swing at smart glasses, and this time it is betting that AI will make the difference. More than a decade after the Google Glass, the company has announced plans to launch AI-powered glasses in 2026. It is creating an audio-only model that lets users speak directly with Gemini. It's also working on a version with an in-lens display capable of showing navigation prompts and real-time translation. The hardware will be developed through partnerships with Samsung, Gentle Monster, and Warby Parker. So, here to help us unpack what this announcement means for Google and smart glasses and wearables at large, we are speaking with Alex Heath, author of the Sources newsletter and co-host of the Access podcast.

17:00Alex, good to see you. Good to see you, man. So these smart glasses, I think the first question is, Google already did it. We had the Google Glass. It was the first example of smart glasses. Then they stopped doing it because it was a dramatic failure. Now they're doing it again. What's changed here? I think the short answer is AI. I think they see a huge opportunity. And I've talked to their leadership there about this to build Google for your face, which honestly is a pretty compelling idea. A lot of them inside Google are also obsessed with this idea of super memory of having all your Google-y context, which is for a lot of people, it's their work, it's their documents, it's their email, it's their calendar, it's maps, it's all these touch points.

17:49Google has like over a dozen products with over 2 billion users, I think. So all these touch points with Gemini as like the orchestrator in a heads up kind of on your face, seeing the world and understanding it, you know, product is a super compelling idea. And that just didn't exist when Glass came out. I mean, AI was but a twinkle on everyone's eye. And society has also gotten to a point where people see the metaglasses out and about. There's all these AI wearables. Sam and Johnny are doing their thing. and there's just this growing, I think, social acceptance of wearables and AI, and I think the timing for them feels good.

18:31And it doesn't hurt that, you know, Meta is seeing early sales traction with the Ray-Ban partnership starting to actually sell, you know, millions of units a year. Those kind of numbers will get anyone who is a potential competitor interested in the space. Do we know anything about what they will look like or how they will work? Do we have any specs out? Yeah, so they're adopting the meta approach, and it's actually the Android approach. I do think they kind of, you could map this to how they've approached Android, where they work with a ton of OEMs. And in this case, the initial products are going to be in partnership with Warby Parker, Gentle Monster, which is a huge eyewear brand in Asia, some others.

19:13And then I do think we will see, I don't know if they're going to call it glass per se, that would be pretty ballsy. I think they should. I've told them that. But we will see a first-party Google experience, too, in the same way that Google has Pixel, but then Android is everywhere, and Samsung sells the most Androids. They really want this platform, this Gemini AR software platform, to be the bet and to be everywhere, which is a real assault on Meta's strategy. Luckily, Meta has the Elisor Luxottica partnership, but Google is going for all the others. And then that just leaves Apple with a fully, you know, classic Apple vertically integrated approach.

19:55And I expect to see, you know, the Apple glasses in a couple of years. It's so interesting to see how this market has evolved over time in Silicon Valley, because, you know, we had one era where everyone thought glasses were the next thing. And we saw the Google Glass. We saw the Snap spectacles. we've even seen you know the vision pro we've seen what's happened with the headsets and all of that then it seems to get put to bed and people seem to say okay this isn't going to happen now it's back and i wonder why it's back is it i mean you mentioned ai is playing a piece of this is it that meta also proved that it can work yeah if i guess they look good yeah if they look good I think, you know, people in tech don't understand this basic concept of like people don't want to wear ugly things.

20:45And I think for a long time it was really hard to get through everyone's head that like the bar to putting something on you with technology is just super, super high. It's not like a phone. It's not like trying a foldable or something. And Meta finally cracked that with Ray-Ban and also Luxottica and a product that's not bad. You know, it's not amazing. The AI is not really what people use it for, but for taking photos, for listening to podcasts, all that stuff, it's a good product. And it looks like glasses, and you can't really tell the difference. And that is the bar. So showing that and then seeing the early sales traction, yes, I think that kind of quote-unquote validates the space.

21:22I mean, man, I've been reporting on air glasses for like a decade, and they were kind of the original AI in the sense of like all these founders all the way up from like Zuck down to Evan Spiegel would talk about this thing as the next major platform that's going to change everything and re-architect how we use technology. And AI ended up kind of being that thing, right, in the last few years. And now I think everyone who's been, you know, the builders in Silicon Valley who've been obsessed with air glasses as a replacement for the phone are finally going like, oh, AI now supercharges this and we have a path to making this socially acceptable.

21:58Just looking at Wall Street's reaction, we didn't really see any reaction in the stock. We did see a huge reaction in Warby Parker stock up 24%. I mean, you know, this is a far smaller company, so that's why we're seeing that. But what are you... That's crazy to me, because this Warby Parker thing was... This is over a year old. Like, they announced they were doing something with Warby Parker, I think, in the spring. Yeah. It just seems a little twitchy to me. Yeah. What do you make of Wall Street's reaction overall? I mean, just on the Google side, Wall Street doesn't seem to care that much about this.

22:32They're certainly not really pricing this in. If this is going to be the next frontier, the next platform, investors don't seem to believe it or take it seriously yet. It's too uncertain. I mean, the timelines are uncertain. Everyone would tell you, they'd tell me in interviews that AR glasses are right around the corner. They're just going to be in three years. And they've been saying that for the last 10 years. So it's just a platform that has constantly been delayed. So I'm not surprised by the reaction or lack of with Google, but it's an AI story, right? There's Gemini involved. So if any Gemini is going to touch your stock as a smaller company, you're going to get that bubble pop that you talk about so much on this show.

23:12Those kind of stock reactions are the bubble, Ed. I mean, let's be honest, like that, that this announcement does not warrant a 24 % jump in Warby Parker stock. I am not a stock analyst, but I just know the, I know, I know facts and I know just like reality and it's ridiculous. It's cause, and so I think just everyone is so twitchy about AI. And if you're, can say you're aligning with who is seen right now as kind of objectively the category leader in AI, Google, uh, of course you're going to get that pop. It doesn't mean it's here to stay, but you'll get it. Just as someone who's covered this stuff and studied these technologies for a long time, what do you think of AI glasses and smart glasses?

23:55Do you think that they are the future? Will we be walking around interacting with screens in front of our eyes? I always have. I've loved the idea of it, the sci-fi idea of it. And I've used so many prototypes over the years. Some are good, some are bad. did you get to use the the meta displays that they just teased or i guess released a few months no i i've tried on i've tried on the the ray bands and taken some photos and yeah played music i've got those um those are a genuinely wow moment when you do the demo um they're not like a product in the sense that they're ready to be like worn all day and used all day and the air is not quite there but when you see the interaction modality the way it like feels the like kind of aha of like oh, getting this context in my vision, it's just like obvious.

24:45It's just obvious, man, that this is where it's going because it's just so much more natural. It's just a natural way to use technology. Again, it doesn't look natural right now and it's clunky and like they've got to iron the bugs out and there's a lot of those. I'm not actually of the opinion that a lot of other, I would say like armchair experts are that like there's all these fundamental breakthroughs that need to happen for AR glasses to be mainstream. I actually do think it's kind of inevitable at this point when you use the meta displays. It's like, wow, they've got a couple years of iterating on this, and this is a pretty compelling product.