Earlier this month, a broad bipartisan majority in the House passed legislation that would force TikTok’s Chinese parent company to either sell the app or have it banned in the United States. A platform once known primarily for goofy dances and launching Lil Nas X to fame now stands at the center of a geopolitical struggle. But what is perhaps even more surprising is how the pending legislation has scrambled domestic political alliances. On the Republican side, all but fifteen members voted in favor of the bill, despite Trump’s vocal opposition, while on the Democratic side, members favored the bill by a 3-1 margin, with prominent progressive voices staking out opposing sides. To help us make sense of the situation, and decide what to make of the proposed ban, the LPE Blog invited six tech and regulatory experts to share their initial reactions.
Ganesh Sitaraman
The TikTok debate has suffered from a lack of historical perspective with respect to foreign-ownership and influence over platforms. Tech neoliberals object to any restrictions on foreign tech platforms because they would constrain the open internet. National security technocrats want case-by-case considerations that are narrowly drawn. As I describe in a paper in the Stanford Law Review, these views are deeply flawed on their own terms. But they also miss the history of how foreign influence over platforms has been treated.
Restrictions on foreign ownership, control, and influence over utility-like businesses have been longstanding in American law, dating back to the founding generation. In critical sectors—banking, communications, energy, transportation law—restrictions have been common, and many still apply today. These sectors were understood to be different from ordinary businesses because they offered platform, utility, or infrastructural services. Such services raise political-economy concerns that are different from ordinary businesses: Power in these sectors can bring a halt to commerce, be used to pick winners and losers, and can shape many downstream activities. As a result, Americans long understood that a different set of rules was needed to govern these businesses.
Tech platforms fit comfortably within this analysis. As I describe in the paper, there are many ways to design a platform-utilities approach to the regulation of foreign platforms. But the key lesson is that there’s nothing new about doing so.
Sanjay Jolly
There are good reasons to be concerned about social media platforms, including the diffusion of misinformation, lack of consumer data privacy, and effects on teen mental health. Of course, none of these harms are unique to TikTok — whose American user base still lags far behind competitors such as Alphabet-owned YouTube and Meta-owned Facebook and Instagram — nor does the pending legislation pretend to address them. Instead, this rare bipartisan intervention into the social media economy is justified not by consumer protection but by national security. Why, then, are American elites so convinced that the presence of a China-based platform in the digital public sphere so gravely threatens American security interests? The answer, I suggest, lies in the shifting geopolitical logic of American statecraft.
American tech companies once embraced China for its cheap labor pool and irresistible consumer market. But starting about a decade ago, as Chinese technologies advanced to compete with American dominance in key sectors, prominent Silicon Valley leaders — Eric Schmidt and Peter Thiel chief among them — began imploring their peers to sing a more patriotic tune. Their purpose was to enlist greater U.S. government support for the domestic tech sector, both through indirect financial subsidies and by aligning American foreign policy to industry imperatives, under the banner of maintaining American technological hegemony. Like generations of tycoons before them, they urged that we accept the private interests of American corporations as the embodiment of the national interest. Coinciding with the American tech industry’s maturation into a mainstay of federal defense contracting, their primary strategy has been to present themselves as the bulwark of the American “way of life” (not only in the United States but across the world) against Chinese rivals who pose an inherent threat to democratic freedoms.
We have seen this ideological blueprint play out repeatedly in recent years. It underlied the proposition that the Justice Department ought to put aside antitrust enforcement in favor of bolstering our “national champions” against Chinese competitors, justified American sanctions against the Chinese tech giant Huawei, defined the State Department’s “Clean Network” initiative seeking to rid global telecommunications of ostensibly “dirty” Chinese technologies, and more recently helped pass the Biden administration’s CHIPS and Science Act. Elite political discourse today accepts the “Chinese threat” as a taken-for-granted fact, regardless of any actual evidence of concrete harms. As a new vision of industrial policy emerges buttressed by economic nationalism, protecting the supremacy of American tech giants becomes a tacit pillar of the national welfare.
Some progressive analysts, meanwhile, have suggested that the bill ought to be understood as an exercise of popular sovereignty. They argue that by preventing foreign ownership over social media platforms, the bill follows a long tradition of public utility regulation that promotes democratic self-government over our communication infrastructures. I see nothing in this legislation, however, that moves us toward treating social media platforms like public utilities — because that’s simply not what the bill does. It imposes no utility-style public interest duties on the platforms, nor does it even really express the proposition that limiting platform ownership to American companies generally serves democratic interests. Instead, the bill targets a specific foreign-based corporation without any showing of its unique dangers, other than the fact that it competes with American firms. Now, it may be that I lack sufficient solidarity with my American compatriots, but I fail to see how democratic self-determination wins the day by granting the executive branch expansive national security discretion over public communications or by divesting TikTok into the hands of Steve Mnuchin or Bobby Kotick.
Indeed, I would enthusiastically welcome a conversation about how we could promote democratic sovereignty over digital communication infrastructures. We should even have a debate about the merits of protecting American digital monopolies from global competition. Such a debate would allow us to probe what Silicon Valley might concede to the American public in return for such protectionism — real data security protections, for instance. So long as “yellow peril” anxieties dictate economic statecraft, however, all the American public will get is a smaller and just as exploitative public sphere, while the world’s richest men entrench their economic fates into the national security state. But that’s the whole point, isn’t it?
Zephyr Teachout
Congress is right to be concerned that Beijing could use the extraordinarily granular data TikTok has amassed on more than a hundred million Americans to identify weaknesses in the American defense systems, and to push targeted propaganda. A divestment bill makes sense – and while I’m sympathetic to the notion that the scope of the bill could be tightened and clarified, other arguments are not so persuasive.
A lot of the discourse around this legislation is about what it doesn’t do – it doesn’t regulate tech privacy, or address algorithmic targeting, or prevent the exploitation of our data. It doesn’t protect us from Meta using its platform power to shape domestic elections. But as I argue over in the Atlantic, this criticism misses the point. We should instead focus on what the bill does: prevent control over a dominant communication platform by a hostile foreign superpower, while beginning to re-build the basic muscle of tech regulation. These are important goals, and the decision about whether or not to support this legislation should be based on whether it accomplishes them. Moreover, the focus on China is not misplaced; according to reports filed under the U.S. Foreign Agents Registration Act (FARA), it has spent more to influence U.S. politics than any other foreign country. The government has shown an interest in local, as well as federal races, while buying up interests in Chinese language media around the country.
I’m also troubled by the argument that the First Amendment could disable the legislation, not because I think a First Amendment argument would succeed (it almost surely wouldn’t), but I’m concerned that it shows an increasing tendency to try to use First Amendment rhetoric to defend big tech platforms. If users have a Constitutional right to use TikTok as it is currently constituted, then tech companies can effectively organize their corporate form and technology in any way they desire, regulating us instead of being regulated by us.
Finally, inasmuch as there is a whiff of cosmopolitan weariness about the nation state itself in these debates, it would be well to remember that the opposite of the nation state is not freedom and equality, but economic globalization.
Nikolas Guggenberger
Legislators supporting the TikTok ban have provided two rationales for the bill: reducing the national security threat associated with data flows to China and limiting the spread of propaganda from the Chinese government. It is dubious whether the proposed legislation will remedy the former in any meaningful way, for the simple reason China doesn’t need TikTok to access data on Americans. A wide range of American companies engage in mass surveillance—extracting information about everything from our commuting routine to our mood swings and fitness progress, which they use to make money in various ways. Data brokers aggregate this information and offer data sets and analytical insights to their customers. If the Chinese government wants data about Americans, they do not need to tap a social media company. They can simply buy that data or obtain whatever analysis they wish. Forcing divestiture or even banning TikTok or similar apps will not remedy the perceived national security risk.
H.R. 7520, which unanimously(!) passed the House largely in the shadow of the TikTok ban, gets closer to drying out data flows about Americans to China. If enacted, the bill would make it unlawful for data brokers to share sensitive personal information entities controlled by foreign adversaries, including North Korea, China, Russia, and Iran. Yet, loopholes remain. First, it is unclear how these limitations could be effectively enforced. How would they, for example, prevent data brokers domiciled in countries which are not on the list of foreign adversaries from obtaining sensitive information about Americans and sharing that information with foreign adversaries? How will US authorities investigate the ownership structures of thousands of data brokers abroad? Second, limiting privacy protections to (the admittedly broadly defined category of) “sensitive data” builds on a naïve narrative that ignores the power of inferences from seemingly trivial data points. The more promising way to protect national security is for Congress to pass substantive privacy regulation, including strict limitations on data collection, data usage, and data transfers in general. Data that is not collected cannot be abused, whether by domestic or foreign entities. And like the EU, the US can then give its privacy protections global reach by broadly defining the geographic scope and by requiring equivalent protections as a condition for international data transfers.
With respect to the propaganda, the proposed ban might prove more promising. After all, propaganda works best if one controls the media infrastructure and has direct access to American users. The infrastructure here crucially includes TikTok’s powerful algorithms. Assuming that the Chinese government had influence over TikTok, a ban or divestiture would take away control and cap direct access to American users. However, again, there might be workarounds. Recall that past misinformation and manipulation campaigns, orchestrated by Cambridge Analytica, Russian Troll Farms, or Macedonian teenagers did not build their own media infrastructure. Instead, they leveraged Facebook—a company as American as American pie. The argument for a ban or divestiture, thus, shrinks from stopping foreign propaganda to inserting some friction into the propaganda machinery.
And here again, there exist more effective alternatives to address this problem: meaningful platform regulation, changes to the underlying business model, and more pluralistic media arrangements. Regulation could rein in algorithmic amplification and establish stricter liability. Changes to the business model, whether through bans on targeted advertising or Pigouvian taxes, for example, could mitigate current incentives to stoke conflicts and spread misinformation. More pluralistic media arrangements would presumably improve content moderation and, most importantly, cut off single points of influence. Considering these alternatives, a ban or forced divestiture looks awfully like mere trade policy that seeks to prop up the domestic surveillance industry in light of successful international competition.
Anupam Chander
In 2020, then President Donald Trump sought to ban TikTok after a summer of being humiliated on the app. He was humiliated first by comedienne Sarah Cooper, who used his own words to ridicule him, and then by K-pop fans who organized via the app to foil Trump’s rally in Tulsa by reserving seats that they had no intention to fill. Four years later, members of Congress, “spurred” by “concerns about Israel-Hamas videos on TikTok,” moved to force a TikTok sale or ban. Many in Congress concluded that TikTok was steering users against Israel in the current conflict, “brainwashing” them. Never mind the reality that rival platforms hailing from the U.S. also show a similar imbalance in their videos on that conflict.
But whether it’s Trump or Biden calling for the sale or ban of a speech app, we should worry when politicians target a foreign-owned speech platform because of the content it carries. Governments across the world have often targeted unfavorable media enterprises, alleging malign foreign influences. We should also be alarmed by the singular focus on the “Chinese Communist Party” as the intruder into U.S. politics. Despite living through Russia’s well-documented election meddling in 2016 and being aligned with a country that is literally fighting a war with Russia, we have not banned Kaspersky’s anti-virus software or Yandex code libraries despite their Russian origin. And while the TikTok bill does include Russia among the adversary nations that it allows the Administration to target, it only focuses on social media, not such things as anti-virus software or code libraries, though the latter could just as easily be utilized for spying purposes.
Note also that the Biden Administration already holds the power to force ByteDance to divest or shutter the app. The Committee on Foreign Investment in the United States could order this now, but has not yet taken that step. Perhaps the Committee is concerned that such an order will fail in court because TikTok offered extraordinary risk mitigation measures, including an elaborate and expensive security enclave for U.S. data, monitoring of the algorithm by third parties, and control over the data and algorithm by U.S.-based directors approved by the U.S. government.
It is perfectly reasonable to worry about online platforms’ impact on discourse or their ability to gather information to share with foreign—or local—governments. However, let us regulate with the industry in mind, rather than a particular company or country. Let us maintain our openness, especially with respect to information. Do we want a world where each country uses only that country’s apps and information services?
Elettra Bietti
The House has recently approved two bills that target “foreign adversaries”: the ban, or divestment, of TikTok from its Chinese parent ByteDance (H.R. 7521), and a (less discussed) Privacy Bill that prohibits sales of Americans’ data to “foreign adversaries” (H.R. 7520). Both are celebrated by a section of progressives who see them as a justified first step toward more comprehensive regulation of Big Tech. Such enthusiasm misses that these reforms conflict with ongoing antimonopoly and antitrust efforts aimed at reining in US tech companies’ power, whose executives and shareholders will benefit from less competition and from the opportunity to invest in TikTok.
“Antimonopoly” or “industrial policy” legislation for the tech industry are empty slogans if all US political elites are able to agree on are prohibitions on “foreign adversaries” interfering with Americans’ digital experience (in a context where, it should be said, the US interferes with the digital experience of every other population on the planet). The two bills in question do not meaningfully protect Americans’ speech rights or their privacy. They do not protect Americans against abuses by their own government or by US firms; and they won’t affect US firms’ surveillance intensive activities or their addictive, harmful and distortive role as information intermediaries. They also won’t change the US government’s ability to co-opt US platform companies into (often racist and xenophobic) crime prevention, counterterrorism, and propaganda efforts affecting the most vulnerable.
It is striking, to say the least, that so many more meritorious reform proposals pertaining to equitable digital privacy, content moderation, and antitrust policy – ones which had a real chance to address tech platforms’ power – never made it through the House. A notable example is the American Innovation and Choice Online Act (AICOA), a pro-competitive Bill that would have imposed obligations on “covered platforms” including Google, Meta and TikTok to refrain from engaging in self-preferencing behavior against competitive threats and nascent competitors.
The AICOA bill was blocked in Congress and criticized by some free speech scholars on the ground that it would help conservative platforms such as TruthSocial gain access to and interoperate with Meta’s democratic user-base. Ironically, if the bill had succeeded in Congress, it would have given TikTok users robust free speech protections against the obsolescence of their accounts following a ban. These users, thanks to AICOA, could have tapped into the interoperable fediverse to keep exercising their rights to free expression. They could possibly port their accounts, content and followers to a different provider or could simply keep communicating with people on other platforms. As Alan Rozenstein has demonstrated, the fediverse can go hand-in-hand with robust content moderation rules.
Yet Americans did not get AICOA. Instead, they might get these two bills. If Americans’ rights to free expression and privacy are to mean anything, they must entail protections against the government and against foreign and US data and speech intermediaries. The selective approach taken up by these two bills, by contrast, does not afford Americans any meaningful speech and privacy protections all.
