By Andrew Stuttaford
Tuesday, December 30, 2025
How quickly time passes.
Just a year or so ago Thierry Breton was, he imagined, a
colossus, the EU Commissioner in charge of implementing the bloc’s Digital
Services Act (DSA). Now he has been barred from the U.S.
The DSA is one of two pieces of legislation designed to provide a regulatory
framework for the provision of digital services throughout the EU. It is
wide-ranging. Some of its provisions are reasonable, others not. One of its
aims was to provide a comprehensive EU-wide framework for the different online
censorship regimes that were springing up in many member-states (with imitators
elsewhere, such as in the U.K.). Hillary Clinton was a fan:
“For too long,
tech platforms have amplified disinformation and extremism with no
accountability. The EU is poised to do something about it.”
Writing about the DSA just over three years ago, I noted
that:
The maximum fine
for breach of the act could be as much as six percent of the previous year’s annual turnover, a sum
designed to ensure that those subject to the law err on the side of caution, as
is the provision that repeated serious breaches of the law could lead to a ban.
Just in case
that’s not enough to do the trick, [the new law] also opens up the possibility
of civil litigation by aggrieved parties, raising the prospect
both of damages and endless lawfare. The DSA applies to all online
intermediaries wherever they are located if they are offering their services
within the EU. That will include Twitter. The EU Commission (the EU’s bureaucratic arm) will
be the primary regulator for “very large online platforms” (meaning those
reaching 45 million users or more), which, in this respect, will have powers
similar to those enjoyed by the EU’s antitrust enforcers, another incentive to
err on the side of the caution.
Material that is
harmful, but not illegal, should not, the EU Commission claims, “be treated in
the same way as illegal content.”
But:
At the same time,
the DSA regulates very large online platforms’ and very large online search
engines responsibilities when it comes to systemic issues such as
disinformation, hoaxes and manipulation during pandemics, harms to vulnerable
groups and other emerging societal harms. Following their designation by the
Commission as very large online platforms [such as Twitter] and very large
online search engines that reach 45 million users will have to perform an
annual risk assessment and take corresponding risk mitigation measures stemming
from the design and use of their service. Any such measures will need to be
carefully balanced against restrictions of freedom of expression.
“Carefully
balanced.”
From the moment that Elon Musk bought Twitter it was
obvious that the company would be in the sights of a Brussels establishment
dominated by radicalized “centrists” who were no friends of either
American-style free speech or, for that matter, almost anything that Musk represented.
What Brussels wanted was for Musk to agree to a content
“moderation” regime that would ensure that the delete button was applied to
“harmful” as well as illegal content, if it fell sufficiently short of
standards demanded by approved, ideologically congenial moderators. Censorship
fans like to claim that, as France’s President Emmanuel Macron put it, the DSA
is merely intended “to enforce online the rules that already apply offline.” But,
as so often with Macron, and as so often with the EU, that is not the whole
story.
As I wrote in an article for National Review in 2024:
What is illegal
under the law of an individual EU member-state or under EU law will remain
illegal. Any amendments to legislation in that area will be left to national
parliaments or to the EU’s legislative process.
But:
[T]he DSA’s broad
language could easily be used to impose de facto censorship on all sorts of
theoretically legal speech, in the interest of preventing “harms” that exist
only in the progressive imagination and that are hinted at in, among other
places, the law’s preamble, but also elsewhere. Thus on its website the EU
Commission warns of the dangers of “climate disinformation.” Tackling that is,
it states, incorporated within its general approach to disinformation,
including making it “more difficult for disinformation actors to misuse online
platforms.”
If
Twitter/X installed the sort of moderators who agreed with the Brussels view of
what such “harms” were, the chance that comments that strayed too far from the
party line remained posted for long — if at all — would be remote. And if X did
not install such moderators, it risked being in breach of the
Act. Speech would thus be banned without formal proscription. Franz Kafka
raises an eyebrow.
Just before Christmas, Ursula von der Leyen, the EU
Commission’s president (its top bureaucrat) tweeted
this:
Freedom of speech
is the foundation of our strong and vibrant European democracy. We are proud of
it. We will protect it. Because the @EU_Commission is the guardian of our values.
George Orwell laughs.
That “climate disinformation” was one area singled out
for this treatment is revealing. One reason for the increasing willingness of
the EU and many European governments to resort to censorship is that the online
opposition (and not only online opposition) to three key pillars of the modern
European order, multiculturalism, mass immigration, and the futile pursuit
of net zero by 2050 is becoming so loud that those in charge in Brussels,
London (what Brexit?), Berlin, Paris, and elsewhere have decided that the only
acceptable response is to muzzle it.
As time passed, Musk and the EU drifted through a phony
peace to a war of words to something more serious. In July 2024, the Commission launched an initial legal
assault on X, accusing it of breaches of the DSA. These were notionally
unrelated to censorship. For example, by switching from its old opaque blue
check “verification” process to a new one that required payment, X had allegedly reduced “users’ ability to make free and informed
decisions about the authenticity of the accounts and the content they interact
with… There is evidence of motivated malicious actors abusing the ‘verified
account’ to deceive users.”
There were other charges too.
There is a Russian phrase, which is usually attributed to
Andrei Vyshinsky, the most notorious of Joseph Stalin’s prosecutors, and is
often rendered as “show me the man and I’ll show you the crime.” If a
determined state is set on punishing someone, there will always be something
for which he can be convicted. The DSA’s extensive and sometimes loosely drawn
provisions gave Brussels room to do just that, and Musk, who was either too
relaxed about the DSA or insufficiently aware of how the Commission operated,
made its task easier than it should have been. X’s alleged breaches of the DSA
allowed Brussels to pretend that this case had nothing to do with free speech
when the reverse was true.
Then Breton let the gag out of the bag. In August 2024,
he wrote to Musk shortly ahead of a talk with Trump scheduled to be
livestreamed on X to warn him to watch what was said as it would “be accessible to users in the EU.” In other words, a
public conversation in the U.S. between an American citizen and an American
presidential candidate was subject to vetting by the EU’s regulators:
We are monitoring
the potential risks in the EU associated with the dissemination of content that
may incite violence, hate and racism in conjunction with major political – or
societal – events around the world, including debates and interviews in the context
of elections… I therefore urge you to promptly ensure the effectiveness of your
systems and to report measures taken to my team…”
Linda Yaccarino, X’s then-CEO, posted
that this was “an unprecedented attempt to stretch a law intended to apply in
Europe to political activities in the U.S. It also patronizes European
citizens, suggesting they are incapable of listening to a conversation and
drawing their own conclusions.”
But Brussels does not want “European citizens” to draw
“their own conclusions.” The EU that had
repeatedly treated inconvenient referendum results with disdain and had now
introduced a censorship framework ripe with potential restrictions is not an
institution comfortable with its subjects’ thinking things through for
themselves.
That said, within weeks Breton was out of a job. His
letter, which was too blunt too soon, was not the reason, but it was yet
another reminder to von der Leyen, with whom Breton already had a poor
relationship, that he was not a team player. On learning that he was not going
to be given a sufficiently senior role in the next Commission (which was then
under formation), Breton quit.
“It’s a good day for free speech,” tweeted
Yaccarino.
Yes and no: Brussels’ machinery ground on, but more
discreetly. There was still speech to be stifled, and another American company
to be looted. Strict comparisons between the fines imposed by the EU on U.S.
companies can be misleading, but the findings contained in a July 2025 report by the (U.S.-based) Computer and
Communications Industry Association (CCIA) are … striking. They show that the
EU and its member-states together fined U.S. companies a total of $11.46
billion between January 2021 and May 2025. Over the same period, American
authorities fined EU companies $4.23 billion, including $0.02 billion on EU
tech companies, the latter a pinprick when compared with the $10.57 billion in
fines imposed on U.S. tech by the Brussels bloc.
In early December 2024, X was fined 120 million euros
($140 million) for breach of the rules alluded to above. Reuters reported that the Commission “said its laws do not target
any nationality and that it is merely defending its digital and democratic
standards, which usually serve as the benchmark for the rest of the world.”
Okey dokey.
Reuters:
The European
Commission’s tech chief Henna Virkkunen said X’s modest fine was proportionate
and calculated based on the nature of the infringements, their gravity in terms
of affected EU users and their duration.
Modest?
It would be interesting to know the basis on which those
fines were calculated. Blue check trauma must be a real scourge.
Virkkunen:
“We are not here
to impose the highest fines. We are here to make sure that our digital
legislation is enforced and if you comply with our rules, you don’t get the
fine. And it’s as simple as that… I think it’s very important to underline that
DSA is having nothing to do with censorship.”
Part of the supposed justification for the DSA is the
fight against disinformation. Virkkunen should report herself to her own
Commission.
X had 60-90 days to file plans detailing how it will
remedy the breaches of the DSA. Given that Musk has described the fine as
“bullsh*t,” this seems unlikely to occur. This will set in motion demands for
even higher payment. X can also appeal in the courts.
Shortly before the EU fine was unveiled, the Trump
administration announced that the resumes or LinkedIn profiles of H-1B
applicants should be reviewed to see if they had been working in areas of
speech policing such as “fact checking” and online content moderation. If an
“applicant was responsible for, or complicit in, censorship or attempted
censorship of protected expression in the United States” he or she should be
ineligible for a H1-B visa. Good.
To its credit the administration did not stop there. On
December 24, Secretary of State Marco Rubio announced that Breton and four other
individuals “who have led organized efforts to coerce American platforms to
censor, demonetize, and suppress American viewpoints they oppose” would be
barred from entry into the U.S. Rubio referred to “radical activists and weaponized
NGOs” who had “advanced censorship crackdowns by foreign states—in each case
targeting American speakers and American companies.” The individuals involved
were “agents of the global censorship-industrial complex,” language that is
less hyperbolic than it sounds. There is a flourishing censorship
ecosystem, much of it dating back to the disinformation panics that began in the last decade, and
quite a bit of it funded and/or used by foreign state or parastatal actors.
Many of those active in this ecosystem cannot, therefore, be seen purely as
private citizens. Rubio explained that “President Trump has been clear that his
America First foreign policy rejects violations of American sovereignty.
Extraterritorial overreach by foreign censors targeting American speech is no
exception.”
Ultimately, this debate concerns a clash of
sovereignties, brought about by the way that social media is domiciled in one
place but available in many others. It is also interactive in a way that radio
is not, and public in a way that phone conversations are not (meant to be). The
EU does not attempt to regulate what is broadcast by the BBC, even though some
of its output can be heard within its realm, nor does it micromanage the way
that British telephone operators are run, even though people in the EU can use
their services to make or receive calls.
The EU’s claim that the visa ban is an attack on its
sovereignty is nonsense. The sanctions amount only to the revocation or refusal
of visas to visit or stay in the U.S., something that is clearly a matter for
any sovereign nation to decide. They come without Magnitsky-style
financial measures. One of those sanctioned, Imran Ahmed of the
Anglo-American Center for
Countering Digital Hate (more on that organization here) has a green card (he is a British citizen) and an
American wife and child. A court battle over his residency appears to be in its early stages.
The other four affected by the sanctions merely face more limited vacation
options. Yes, the administration’s move is designed to put pressure to those
who would censor American speech, but if the EU believes this is bullying, it
should take a look at its own behavior.
But there is more to the sovereignty question than that.
Nations are legally entitled to police speech within their own borders. That so
many Western democracies are doing so in an increasingly oppressive fashion is
appalling (and a testament to the importance of the First Amendment), but if
their voters truly want to change that they can vote to do so (exceptions or
exclusions apply within the EU, an institution moving on to post-democracy).
Put that way, matters are straightforward. American social media companies
doing business in those countries should comply with local laws. But what, if
in doing so, they infringe on Americans’ free speech rights within the U.S.?
If X starts subjecting communications between Americans in America to scrutiny
by censors in Brussels, London, and elsewhere, there would, quite rightly, be
an uproar.
However, for Musk to ignore the EU’s fine and carry on as
before could, at least in theory, prove ever more expensive, as the penalties
for noncompliance pile up. Moreover, the Commission is investigating other
areas of X’s business. Fresh fines, almost inevitably, will be on the way. As a
practical matter, unless X holds large assets outside the U.S. (I don’t think
it does), it would probably be difficult for the EU to enforce any judgment
against X, although for the company to be paid by its EU customers would be
tricky. Musk’s travels in future might also become very circumscribed. And this
is before considering censorship regimes such as those in the U.K., which
include criminal penalties for managements that do not knuckle under.
The U.S. should use its bully pulpit to encourage its
allies to ease up on their online censorship and it should do more than that to
take action — sanctions on more people, to start with — against those taking
action against American companies acting as a platform for American speech in
America. Such efforts, however, would be unlikely to yield results any time
soon, leaving Musk with a set of unlovely options. Assuming that he will not go
along with censoring American speech, he can, as mentioned above, proceed as
before, with all the difficulties that might entail. Alternatively, he can
simply block access to X by accounts based in the West’s censorship states. But
so long as this does not provoke a voter revolt, this would be to hand victory
to the censors. Brussels, well aware of how it would look and of the legal
obstacles involved, is unlikely to follow the lead set by countries such as
Russia, China, and Iran and block X itself.
Writing in November, Breton envisaged a world of
“state-controlled digital empires,” American, Russian, Chinese, and European,
each reflecting separate “vision[s] of the information space, shaped by their
values, priorities and their relationship to the market and the state.” The
nature of the Russian and Chinese digital empires is not hard to guess. That of
the Americans, maintained Breton, would be “built on the primacy of private
actors and minimal oversight.” Let’s hope! By contrast, Europe’s would be
designed to “ensure cohesion, protect users, guarantee transparency and
safeguard the foundations of our democracies.” “Cohesion” in this context is
code for soft authoritarianism, and “protection” is, in most cases, a synonym
for the infantilization of the general population. The talk of “transparency”
and “democracy” is too obviously self-serving and dishonest to merit serious
discussion.
One alternative might be for X to expurgate content
accessible by users within the EU, U.K., and other censorship-heavy
jurisdictions. Technologically such a step is straightforward enough (broadly
speaking, this is what it already does with respect to users in Germany and
some other countries), although it could be expensive to administer and it
would involve X not only collaborating with the censors but remaining subject
to their second-guessing and the fines that could go with it. One plus is that
the way that such censorship would work is that users would be told that a post
is inaccessible because of restrictions in their part of the world. Seeing
enough notices like that could be a useful wake-up call. Maybe. In the
meantime, the determined could peer through the Brussels wall with the help of
a VPN.
For as long as they are legal and allowed to function as intended, that is.
No comments:
Post a Comment