Home

A Recap of the US-EU Trade and Technology Council Meeting with Mark Scott

Justin Hendrix / Jun 4, 2023

Audio of this conversation is available via your favorite podcast service.

Last week, the U.S Secretaries of State and Commerce and trade representatives from President Joe Biden’s administration met with top European Union officials in the heart of the Swedish Lapland for the fourth Ministerial meeting of the U.S.-EU Trade and Technology Council, or “TTC”.

Pressing needs were tackled, new initiatives were launched, commitments were made, and cooperation was deepened on a range of tech policy issues, at least according to the press releases.

To hear an unvarnished view from someone who was at the meeting about what might actually come of it all, I invited on a journalist who is, in my opinion, one the best tech policy reporters in the world: Mark Scott, Chief Technology Correspondent for Politico. With his colleagues, Mark filed multiple pieces from Luleå, Sweden, where the event was held, including on the degree to which the specter of China loomed over the meeting; on tough talk from European Union internal market commissioner Thierry Breton; and on the dialogue around artificial intelligence.

What follows is a lightly edited transcript of the discussion.

Justin Hendrix:

Mark, you have just returned from a very interesting meeting that I feel like is sort of less well covered, a little bit off the beaten path. Can you tell folks just what went on in Sweden and what you observed there?

Mark Scott:

Sure. I mean it wasn't just Sweden, it was northern Sweden. We were just in the south of the Arctic Circle in a place called Luleå, which is a town of about 75,000 people, and forgive my awful pronunciation. It's the biannual EU-US Trade and Tech Council Summit.

What does that mean? It has been going on for about two years and it involves Anthony Blinken, Gina Raimondo, the Commerce Secretary, and Katherine Tai, the US Trade Representative getting together with their European counterparts, Margrethe Vestager, who is the... She has a long title, but basically she's the European Commission's digital chief and her trade counterpart, Valdis Dombrovskis.

They get together, they hash out hopefully thorny issues on trade and tech to create some sort of transatlantic unity. The idea basically is to sort of bridge some of the divides that came up during the Trump administration.

Justin Hendrix:

I think from the US side, it's perhaps right to see this as part of the kind of generally pro-democratic effort that the Biden administration has been on. It's sort of a sort with the Summit for Democracy, the Declaration for the Future of the Internet, some of these types of activities.

Mark Scott:

Yeah. Reading from my notes, "America's back," right. This is all about promoting to European allies that DC is taking these things seriously.

Justin Hendrix:

I appreciated some of your coverage. You included some great details. You talked a little bit about the fact that the AI conversation in particular took place over a lunch of cod loin and chocolate praline. AI seemed to have been the topic du jour. What was it all about? What was the goal between the US and EU counterparts there?

Mark Scott:

Sure. To caveat, I did not get the chocolate praline, so that's a shame from my end. The AI discussion has been going on within the Trade and Tech Council since the beginning, since sort of early last year. But it's been very technical. The idea being that what do we do with this so-called trustworthy AI?

They've been hashing out through very wonky working groups that meet sort of quasi regularly about metrics and measurements and bringing in the best of both sides. That's great and that's kind of what industry wants. They want to make sure that, yes, the Europeans are going a mandatory legal route, the Americans not so much, but where is the common ground? That all changed when ChatGPT came around. We are seeing this in DC, in Brussels, in Brasilia, in New Delhi, wherever you go. It's sort of, "What do we do about generative AI?"

About two, three months ago, the discussion that was very technocratic changed to, "Okay, let's do something, or to be seen to be doing something on generative AI." You had Blinken and Vestager and all the others at the final press conference making a concerted effort to discuss something that they're calling a voluntary code of conduct. We don't know what that means, but basically it's an renewed effort to sort of take a big swing at a macro level at generative AI in particular to say, "Look, we know you guys are... The public's freaking out. We're on this. We're going to do our best to come up with some solutions."

Justin Hendrix:

These codes of conduct or codes of practice are I think of as sort of very European. All about getting industry to agree to certain principles, certain objectives, and then perhaps seeing if that can lead to, I guess, a framework for regulation.

Mark Scott:

I do not want to name-drop because I'm not a very important person, but I got to spend some time with Margrethe Vestager at the end of the conference, and she was basically saying that this is a two-page briefing note that the Commission wrote out themselves, and she personally handed it over to Raimondo over the cod loin and praline lunch. This is a very much currently a European-centric approach to codes of conduct. It's a question of will the Commerce Department and maybe the National Security Council in the White House will they agree to it? We need to figure that out.

We now have a couple of so-called Sherpas on both sides to take the loads on actually what will be involved in this code of conduct. But the idea is to create some sort of transatlantic voluntary code to present at the G7 meeting that's going to be held sometime in the fall.

Justin Hendrix:

We've got this separate process, this G7 AI process, and we've got three dedicated expert groups that are coming out of this that'll focus on, it looks like, AI terminology and taxonomy, cooperation on AI standards and tools for trustworthy AI and risk management and monitoring and measuring existing and emerging AI risks.

This is all from the readout from the White House. I see a list of 65 key AI terms essential to understanding risk-based approaches to AI, which the US and EU are working to harmonize. This all sounds great, right, two of the world's biggest economies working on AI together. But the US and the EU really don't see eye to eye on how to go about this at all.

Mark Scott:

I mean, that's putting it very mildly, right. I think you need to see this as two tracks. The metrics and the measures you mentioned were existing AI work within the Trade and Tech Council that was going on for the last two years. It was very technocratic, and frankly no one cared because it's sort of things that only the boffins of the boffins care about, to use a British term.

Then there's this code of conduct which is like, "Oh crap, we need to be doing something on generative AI now and therefore we're going to bring in the Secretary of State, we're going to bring in Europe's digital chief to give a broader macro view," which they inserted it at the last minute, literally the two-page document discussed on the last day. That is the thing where we need to focus.

But that doesn't get away from the fact that Europe has its AI Act, which they're trying to get done by December. It's supposed to ensure that companies can't use the certain technologies in so-called harmful use cases.

As we've seen in the last sort of month or so, with multiple executive orders and meetings at the White House and even Sam Altman from OpenAI testifying to the Senate, the US is taking an interest, but no legislation is going to get passed this term before 2024. The White House itself is taking, from my opinion, a very industry-led approach to what to do about this.

Justin Hendrix:

Yeah. You had senators kind of talking to Sam Altman, well, by his first name, calling him Sam. They'd all just come off a dinner with him apparently the night before the hearing. So you do get a sense of this closeness. When Biden hosted his summit on AI, it was of course corporate executives that were in the room rather than critics or advocates for reform.

Mark Scott:

I mean that makes sense, though, right? I mean you look at, okay, let's take OpenAI and Microsoft, Google, Meta as maybe four who have arguably the most invested in AI, and they're all American companies. There was a commercial interest to do this, and that's a logical way to go about this.

You then look at what goes on in Europe, and I'm not saying that as sitting here in London, but they have a very different approach to this because there are fewer companies in Europe doing this, and therefore it is seen as a potential, another area where the Americans can dominate. Rightly so, they've invested the most, but it's definitely there is a commercial interest underlying how both sides are looking at this.

Justin Hendrix:

Is there a sense in the conversations you've had either I suppose with some of the leaders or perhaps in the hallway that the US is just going to go this laissez-faire route again, and the same sort of situation that we have with social media is about to repeat itself where American, corporate, dominance, power, essentially these products kind of propagate themselves, and Europe is sort of left to try to figure out how to reign in their practices and maybe try to protect their own citizens within their own principles and frameworks?

Mark Scott:

I mean, I think there are a lot of people in the administration who also want to protect Americans, right. This is not about letting industry do what industry does. But I do think just looking at the political dynamics in the US right now in terms of Congress and even some of the states, it's very difficult to pass any legislation, let alone digital legislation.

I think when it comes to AI, we're having a lot of rhetoric and a lot of executive orders and even DOJ and the FTC saying, "We'll use our powers to do what we can," but there's no federal privacy law, right. There are issues with how this works in the US that makes it more difficult. And as much as I have issues with the European approach on how they go about regulation, they have been working on an AI act for more than two years. The fact that until very recently it didn't even look at generative AI is a problem, but they're trying to fix that.

It is just a question of just how both machines work that it's very different and getting them to both agree on some sort of voluntary code beyond platitudes is going to be difficult.

Justin Hendrix:

AI was not the only subject on the table at this meeting. And I want to kind of just throw out a few others and get your take on what was discussed. Maybe I'll just give you one or two word prompts and see what comes out.

One thing that stood out to me: digital identity.

Mark Scott:

Oh, yeah. This is one of these wonky topics that the Trade and Tech Council is supposed to fix. This is about helping propagate sort of standards and rules across both jurisdictions so they can frankly take on China. I mean, no one wants to mention China on the European side, but it's about China. The idea of trying to share best practices on digital rule and digital ID is crucial to this because it both plays into the trade aspect, but it also allows to sort of digitize tax systems and all the other kind of technocratic things that no one cares about but can be quite effective when you're looking to cut costs and make things more efficient.

Justin Hendrix:

Semiconductors.

Mark Scott:

Ah, the chips to issue. As you know better than anyone, both sides have invested or will be looking to invest billions of dollars to subsidize domestic semiconductor production. The issue with that is you've got the Intels and the TSMCs in the world going around and saying, "Hey DC, Brussels has offered me 10 billion. Can you offer me more?" What they're trying to avoid is a so-called subsidy race.

The semiconductor conversation is basically, "Let's be very transparent about where we're going to spend our money." If Intel comes to me, and other companies can do it too, if Intel comes to say, the US state says, "Can you offer me this much money? Because the Europeans have done it too," they can check to make sure the companies aren't lying. Therefore, you're not ending up saying horse-trading between two jurisdictions who have equally valid reasons to bring semiconductor production home, but there's a opacity on what the negotiations look like. The idea is to sort of share that information at the frontline so that money can be best spent is the practice.

Justin Hendrix:

We've got a brief little mention of quantum.

Mark Scott:

This is about R&D. Arguably, as you said, the US and Europe are two of the, at least the Western world's, biggest democracies and market economies. Quantum, if you're looking to take on China, is a massive issue. How do you then maximize that so people in Toulouse and Tallahassee can share whatever they're doing to maximum effect.

This is about, again, things that... I get why the public don't care that much about, but if you are working in quantum and all of a sudden you can access European Horizon 2020 funding, which is billions of dollars, then that could be quite useful to you and vice versa. If you are a European looking to access sort of, I don't know, commerce money as well.

Justin Hendrix:

You mentioned China, and some of your reporting has been on the specter of China. There's also a segment in here really on the geopolitical context, and one of the things that stands out is coordination around disinformation and what the group calls Foreign Information Manipulation and Interference, FIMI, which is a new acronym that I'll have to start to adopt.

Mark Scott:

That no one will use, to be frank. This is an interesting one. There is an obviously a politicized conversation around misinformation and disinformation right now going on in the US, particularly in the House of Representatives. Part of the discussion has been about let's get into some of the misinformation, foreign interference question, but let's not touch anything domestic because that has domestic political implications for particularly the US side.

What they're doing, and then name checking Russia and China in particular saying you are interfering in Latin America and Africa, so we are going to coordinate our response in terms of pre-bunking, as they call it, getting ahead of Russian disinformation tactics in say Colombia to show like, "No, you're doing this."

The problem with that is when you talk to the people who are doing this work on the ground and say, "Hey look, you get to now work with the DHS and then you get to work with the G7 rapid reaction unit, all these other kind of groups that are working on disinfo," they aren't really involved in this new coordination regard by the Trade and Tech Council. I still remain somewhat skeptical. This is not just, again, platitudes, like what is actually the meat to the bone that we're going to see from this?

Justin Hendrix:

One of the things that really stood out to me was that some of the language around child safety and online concern around mental health and some of those things also made it into the readout. Were you present for any of the discussion about those issues?

Mark Scott:

Biden mentioned in his State of the Union address and also there is particular child safety legislation coming in Europe. Without being massively cynical, and this is an important issue, you don't lose votes by saying, "I want to protect children online," and then it is a massive mental health issue.

There is a concern going on around what platforms are doing and what you've seen with a variety of age appropriate design codes coming out of the US states. This is a big issue coming through, at least at a local level in the US too. I think that is a question of we are recognizing collectively that we need to do a better job of protecting children and what they see online.

Justin Hendrix:

Let's talk about the piece that comes directly after that, which is this concern around enabling independent research. I think you were the first to report that this was a kind of focus of the conversations. I guess the promise was that perhaps there'd be an opportunity to somehow harmonize the independent research access provisions of the Digital Services Act in a way that would make it possible for US researchers to have similar access or perhaps for there'd be some other codification of such access in the United States. Is that something that got discussed? Is that a possibility?

Mark Scott:

I think it is. Let's take a quick step back and say why this is important. We have the US, European, UK and Indian elections next year, and we still have a very limited knowledge base about what's going on on social media. Data access is crucial to this. It provides transparency, accountability. It lets everyone else, journalists included know what's going on. It's really important and companies like Reddit and Twitter are cutting back on this access. It's quite crucial for this to happen. The Digital Services Act in Europe provides mandatory requirements for the platforms to do this, and we're going to see that roll out in the next, say, 6 or 12 months.

The question then becomes if you're offering it to Europe ahead of next year's election, the White House and other parts of the administration have a legitimate question to say, "Well, why don't you offer that to us too?" There's a track record for this back in 2018 when Europe passed its GDPR data protection overhaul, some of those provisions and protections about consent and data usage got transferred over to the US in sort of a, what I call, a GDPR light that allowed the Europe... the US citizenry to benefit from some of the stuff that was going on in Europe. It wasn't perfect, but it was something similar.

What I think what we're going to see hopefully by December, at least that's what the timeframe I was given that some sort of... Again, these are my words, DSA light regime coming, at least announced by the White House or the administration in December that would allow US researchers to have similar-ish voluntary data access from the platforms akin to what is being made mandatory in Europe.

Justin Hendrix:

Do you know anything about the sort of US side of that, and that would essentially be done under some kind of executive order or otherwise, I don't know, administered by a federal agency?

Mark Scott:

I mean that is the $64 million question, and there's been a bit of a churn in the White House in these topics in the last couple of months. Ask me in a month when I hopefully get to speak to some people to know what's going on. But I do think, again, it's very tentative right now, but the language in the Trad and Tech Council communique that came out, which you're reading from, was basically lifted word for word from the Digital Services Act.

And so if we're going to take that at their word, and there is a conscious effort within the White House to do something on this, and the fact that Congress is pretty much shut down for political reasons, these are my best supposition, you are looking at some sort of executive order, or at least arm twisting of the platforms to say, "Can you do this?"

Because what happened with GDPR is the platforms basically did that themselves based on some strong words from the administration. Will we see an executive order or will it be sort of quiet words in a backroom somewhere? It's to be determined?

Justin Hendrix:

If there are any US officials listening to this by any chance that can shed any light, I'm sure I'd like to hear from you, and I'm sure Mark would as well. Let me ask you... Grab bag time. Anything that we've not discussed that you think was important? There was a lot around connectivity, 6G, which some listeners may be surprised to learn is now on the agenda. It feels like we've barely choked down 5G. Lots of connectivity issues, lots of specific issues to do with particular countries and other types of bilateral conversations that are important both to the US and to the EU. What else should we pay attention to coming out of this?

Mark Scott:

It's an interesting one. I think we can't forget the geopolitics of this. Again, that's kind of my jam. I kind of look at the tech from a geopolitical lens, but I think the fact that you have these two close allies despite their sort of so-called frenemy relationship coming together to talk about 6G and telecommunications and semiconductors and all these issues, you can't escape the fact that China is in the room but not, right.

You look at what the US administration, both Biden and Trump, did around Huawei and now they're looking to do around TikTok in terms of both... and semiconductors to a degree of trying to limit China's encroachment into this sort of tech and digital space, that is underlying some of the reason why at least from the US side, you're seeing, let's work together on 60 standards so we can have a Western consensus to push out China. Let's do stuff on semiconductors to push out China. Let's look at some of the other issues from the US side to bring the Europeans on to our way of thinking.

From a European perspective, it's a bit more complicated because sometimes the US doesn't get this, but Europe is in a monolith. There are multiple different jurisdictions and countries with different interests. From the US... European perspective, it's about coaxing the US to take on a more European style regulatory approach to tech. That goes from everything from the platform stuff we discussed even to 60, which frankly the Europeans compared to the US is leading on with Nokia and Ericsson versus maybe Intel on the US side there.

There's sort of a push and pull between the US kind of hoping and willing the Europeans to take on the anti-China hawkish view and then from Brussels, the idea that we're winning this whole Brussels effect by sort of nudging the Americans towards a European style sort of regulatory regime. It's somewhere in between. Neither side is getting everything they want, but the fact that you've got Anthony Blinken, Gina Raimondo, Katherine Tai, and the Europeans coming to a town in the middle of nowhere... It literally is the middle of nowhere... for two days, I think, is emblematic of at least that transatlantic ties being fostered again.

Justin Hendrix:

When you were there in the room, is there a sense of a pecking order? Is there a sense of US throwing its weight around and its sort of innovative muscle or European officials having to, I suppose, advocate for their point of view in this regard? Do you get a sense of that as an observer?

Mark Scott:

I think it's very mutual. I mean, I know that it'd be nice if one was sort of bullying the other because that's the better story, but I do think there's a lot of mutual respect between sort of the mid-career officials who have to sit in endless Zoom calls to hash this stuff out. I think there is a relationship there, and again, it's not about tech issues, but the Trade and Tech Council also looked at export controls on Russia after the war, and then it was very successful to hash out that very quickly because the officials knew each other.

I do think there is a question about what happens next year if there was a change in administration in the US. Does it stick around? Is this just another glorified four year pet projects for Photoshop, for photo opportunities and then it disappears? That is a massive unknown, right. That is a question.

But I do think currently, and this is partly because of a lack of muscle memory in the US side when it comes to digital policymaking, the fact that the Europeans, like it or not, have the Digital Markets Act, the Digital Services Act, the Digital Governance Act, the Data Act, the AI Act and GDPR... Sorry, that's a lot of acronyms. They just have a bigger pool of resources to bring to the table compared to what does the US offer? Don't play nice with China is pretty much the message.

The US has a lot of the good technical skills in NIST and NTIA to help with some of the standardization, but in terms of here is a prescribed rule book, which you can do now, it is a unfortunate fact that the Europeans just have more of that than the US does at the moment.

Justin Hendrix:

You've mentioned this idea, this conversation is very much happening under the specter of China, and that to some extent this is about harmonizing US and EU views. But do you think there's a shared sense of the future between these two delegations of the future that we want, of Silicon Valley's role in the future that we want? Or is it still coming from a very different place?

Mark Scott:

It's still coming from a very different place, and I say that with someone who has a foot in both camps that there is still a regulate first, ask questions later mentality in Europe that doesn't really have any commercial imperative, mostly because there are no companies really here... Sorry, Spotify and SAP... and on the US side because there was a lack of movement in Congress and then in the beltway and because the companies are primarily based in the US, there is a sort of legitimate commercial reason why, again, that we've moved on from this phrase, but sort of move fast and break things mentality still exists to a degree.

I still think there is a wariness from the US to think that everything out of Europe is protectionist and the Europeans still have a misconception that all American policymakers think they're in the pockets of Meta and Amazon. Neither is true, but I still think despite two years of multiple discussions, regular meetings, there is still a question of, 'we have a different approach to this. We don't really have a solution. But if we keep meeting in the middle of nowhere in Sweden every six months, or Pennsylvania or Philadelphia with...' No, excuse me, Pittsburgh, it was indeed, in the US, 'we can figure something out.'

What we see with the voluntary code of conduct on AI and generative AI is a clear example of that's a wonderful headline, but it's a two-page document that has no detail to it, and you have less than six months to formulate that to present to the G7. What happens with it, and if anything happens from it is a clear litmus test both for the Trade and Tech Council but also for the transatlantic relationship.

Justin Hendrix:

Well, perhaps we'll have you back when you've had a chance to go to Hiroshima, and maybe if they didn't give you any cod and pralines, maybe you'll have some nice food at that meeting.

Mark Scott:

I hope so.

Justin Hendrix:

Mark, thank you very much for joining me and hope to talk to you soon.

Mark Scott:

Thank you.

Authors

Justin Hendrix
Justin Hendrix is CEO and Editor of Tech Policy Press, a new nonprofit media venture concerned with the intersection of technology and democracy. Previously, he was Executive Director of NYC Media Lab. He spent over a decade at The Economist in roles including Vice President, Business Development & ...

Topics