Europe, America, and the Digital Economy
Mark Zuckerberg’s very long day in DC and Macron’s WIRED interview provide contrasting approaches to disruptive innovation
Washington
The week of 9 April saw two important case studies on the role of governments in disruptive innovation: the first being Mark Zuckerberg’s series of hearings on Capitol Hill, and the second being French President Emmanuel Macron’s interview in WIRED describing France’s recently-released “AI Strategy”.
Zuckerberg’s Senate hearing was, in my view, largely a disappointment. The first impression I can conjure to mind is that of a massive missed opportunity for Capitol Hill to take a more assertive role in the immense market disruption coming out of Silicon Valley. The Senate struck me as keen to relinquish that responsibility. Most senior Democratic senators honed in obsessively on the ongoing investigation into Russian interference and fake news, while the senators present generally showed ignorance of the basic workings of Facebook and its products. Furthermore, the core questions of the debate over the future of the digital economy were left largely untouched. Zuckerberg was let off easy, considering the lack of seriousness with which Facebook and other tech giants have treated user privacy up to this point. Moreover, the trust-busting spirit of Washington seemed entirely absent, save a few stray comments about market dominance.
For me, the key questions, from a regulatory standpoint, were:
1. Is Facebook a monopoly?
2. Is social media a neutral platform?
3. Is social media fundamentally different from other online services?
A few senators however did head in the right direction on these questions at various points during the marathon hearing. Here’s a list of the senators I’m giving an A+ for having brought up the three questions I listed above.
Question one was touched on briefly by Senator Lindsey Graham (R-SC), but Zuckerberg was allowed to shrug it off with a joke. Senator Graham asked if Facebook was free of competition, to which Zuckerberg answered, “well, it certainly doesn’t feel that way”. Ha ha. This is the fundamental question of the fate of American tech giants: should a single company be responsible for the management of the platform through which most Americans get their news. Such monopolies do not exist in print, television, or radio. Sarah Miller argues in the Daily Beast, radically but at times persuasively, that “Facebook is a corporate monopoly whose very business model is surveillance and user manipulation”, and that it must therefore be broken up for the public good. Whether or not one agrees with this statement, a senatorial hearing was a huge missed opportunity for a public debate on this issue.
Question two has been front-and-center in the public forum since the 2016 election and successive revelations of the propagation via social media of ‘fake news’ sites. Senator John Cornyn (R-TX) did pin Zuckerberg to a half-answer to this question, though details were not discussed. The Senator asked if Facebook was “responsible” for the content shown on its platform, to which Zuckerberg responded authoritatively, “we are responsible for content”. However, a desperately needed follow-up would’ve been “in what way is Facebook responsible for content?” Without further elaboration, this statement remains an empty commitment.
Question three was hinted at by Senator Roger Wicker (R-MS): “should privacy regulations apply to all websites equally, as advocated by ISPs, or should social media companies (“edge providers”) be treated differently based on their unique business model?” Great question. This logic leads to a larger question: is social media just another digital business? Or must it have specific regulations, specifically for its treatment of consumers’ data, given the highly personal and private nature of the insights gained from monitoring billions of individuals’ social lives.
The final senator I’ll mention is Kamala Harris (D-CA), former California Attorney General and frequently-mentioned hopeful for the Democratic nomination for the 2020 Presidential Election. The former prosecutor was one of the most junior senators in the hearing, and thus unfortunately one of the last to speak after hours of questioning. Her opening statement neatly summarizes the concerns raised by the other three senators I’ve mentioned:
“During the course of this hearing, these last four hours, you have been asked several critical questions for which you don’t have answers. And those questions have included whether Facebook can track user’s browsing activity even after the user has logged off of Facebook, whether Facebook can track your activity across devices even when you are not logged into Facebook. Who is Facebook’s biggest competition? […] I’m concerned about how much Facebook values trust and transparency.”
While the above senators’ interventions were on-target, the vast majority of questioning missed the heart of the discussion.
Many Democrats essentially wanted Zuckerberg to admit he made Hillary lose, and Republicans’ questions either missed the fundamental questions at hand or demonstrated a clear lack of understanding of Facebook’s and its subsidiaries’ products. Notable soundbites in this category included concerns over “emailing in WhatsApp” triggering unwanted ads, whether “Facebook platforms talk to each other”, and Senator Blunt (R-MO)’s shout out to his 13-year-old son’s “dedication” to Instagram. Cute.
The Zuckerberg hearing demonstrated that many senators were either unable or unwilling to seriously engage with the long-term implications of disruptive technologies. The public requires a governmental voice which is able and willing to shape debates effecting modern Americans’ interactions with the digital economy, and to protect their interests. From this perspective, last week’s hearing was a missed opportunity. On AI and big data, the government has and will continue to cede territory to big tech. I do not say this because I think the tech industry necessarily needs excessive regulation, but because the nature of the economy online has fundamentally shifted from the regulatory frameworks established for print, radio, and television, and thus the discussion and approach of Congress needs serious revision and updating.
While this lack of engagement will have no impact on America’s capability to compete with the Chinese in AI — American tech giants are far beyond any of their peers — the ignorance demonstrated during the Senate hearing on basic technical subjects has serious implications for American citizens. As Macron told WIRED, “all the [tech giants] will make are private choices that deal with collective values”. Lawmakers are essentially ceding responsibility for the shape of the new economy to actors which are not democratically accountable. American tech giants are responsive to markets. However, these super-corporations hold partial to complete monopolies in their respective fields. They don’t just respond to trends, they generally shape them.
Europe
In Europe, France’s commitment to spend on “public, socially responsible” AI development and the EU’s General Data Protection Regulation (GDPR), which enters into force in under a month, both show a newly assertive Europe which is prepared to engage seriously with the fundamental questions underpinning the individual’s relationship with disruptive innovation and the digital economy.
At the beginning of his mandate, Macron tasked mathematician-turned-parliamentarian Cédric Villani with designing a public approach to artificial intelligence for France. The resulting March 2018 report was entitled “Making Sense of Artificial Intelligence”, totals 235 pages, and outlines a “national and european strategy” for harnessing AI for the public good. “Villani’s AI report and Macron’s WIRED interview discussing its resulting strategy are more focused on machine learning than on AI in general, but the implication is clear: France wants to represent its citizens in the face of the profound changes to the labor market it sees on the horizon.
France’s strategy has six basic points:
- AI is a data-centric technology, so data regulation is key to mastering its effects;
- Research should focus on healthcare, environment, transport, and defense, and an “Interdisciplinary Institute for Artificial Intelligence” should be created by the State;
- The regulatory environment must be prepped for massive changes in the labor market;
- AI can be wielded to help the green transition;
- The State must be ready for AI’s impact on other key fields for its interests, namely law, social services, and the political process;
- AI must expand access to the labor market to make it more “inclusive”.
The problem with the French approach is that it probably won’t work. France’s tendency for state-driven innovation carried out by public and semi-public actors worked great for the TGV and Arianne, but these are heavy industrial projects which rely almost exclusively on steady, long-term government procurement. If the problem facing your domestic technology companies is that they’re not able to keep up with fast-moving American competitors in terms of scale, recruitment, and compensation (RIP Minitel), then the solution is not to replace these actors with a lumbering “interdisciplinary” public research bureaucracy.
I’m by no means an evangelist for the private sector, but AI is a disruptive field in its relative infancy, and these types of industries are simply not well-suited to institutionalisation. The Villani strategy promises “decentralized and agile innovation”, but I remain skeptical. The public-centric approach has worked in areas where profit is scarce but scientific importance is high, such as the CERN facility in France and Switzerland, but such projects are poorly-suited to disruptive consumer technologies like artificial intelligence. CERN might have invented the basis of the World Wide Web, but this innovation didn’t end up in the hands of French or Swiss companies, it ended up being capitalized by more agile, profit-driven American corporations in Silicon Valley.
Moreover, France faces serious budgetary constraints in competing with the Americans and the Chinese. The impressive €1B-plus that the government has announced for AI over the next five years is equal to one-year’s investment in AI by the American government alone, not counting the billions invested every year by American GAFA (Google, Amazon, Facebook, Apple), as well as ride sharing company Uber. China has invested $2.1B in a single project alone, an industrial park for AI outside Beijing.
But don’t despair: the best part of Macron’s push for AI might simply be that he’s doing it at all. To quote Woody Allen, “the most important part of life is showing up.” Europe, by seriously engaging with the impact of disruptive AI-based technologies, is winning a “participation trophy” (ref some joke about millennials) of sorts. Macron himself argued to this effect in his WIRED interview:
“Artificial intelligence will disrupt all the different business models and it’s the next disruption to come. So I want to be part of it. Otherwise I will just be subjected to this disruption without creating jobs in this country.”
In this logic, playing is winning, particularly with an economy the size of Europe’s throwing its weight around. Whatever one may think of GDPR or of Macron’s AI Strategy, leaders in Europe are grappling with these big questions in ways that Capitol Hill, beyond military applications, is not.
The EU’s assertive stance on privacy isn’t limited to data, but also to encryption and communications technologies. Aside from GDPR, the Commission’s consumer watchdog, the “Article 29 Working Group”, recently published an opinion detailing a privacy-centric approach to encryption:
“The availability of strong and efficient encryption is a necessity in order to guarantee the protection of individuals with regard to the confidentiality and integrity of their data which are the elementary underpinning of the digital economy. Backdoors and master keys deprive encryption of its utility and cannot be used in a secure manner.”
By taking an innovative, pro-privacy stance on big issues, Europe has already shown it can shape global policies, such as the extraterritorial application of GDPR principles by Google, Amazon, and now also Facebook. I recently got this notification from Twitter as well:
For a market space like Europe, playing the game is winning. When a diversified market of 500 million consumers with strong institutions throws its weight behind big principles, they can drive change, irregardless of their individual companies’ ability to compete one-on-one with American tech giants. By engaging seriously with issues like privacy, machine learning, and disruptive innovation, European legislators and governments are protecting European companies and consumers from the threats posed by unregulated markets abroad.
Jackson Webster is a graduate of the King’s College of London Department of War Studies, and is currently reading for a master’s in International Security at Sciences Po Paris. His research focuses on Russia, its relationship with Central Europe, and cybersecurity. He is currently working on cybersecurity and data privacy issues with Seville More, a legal tech company in Paris. You can follow him on Twitter @joliverwebster
