The EU Court’s decision is no revolution: it confirms that platforms are not merely passive repositories of content. But the real problem lies elsewhere: entrusting private entities with the task of assessing the lawfulness of content in advance risks curtailing users’ rights and freedom of expression by Andrea Monti – Initially published in Italian by Italian Tech – La Repubblica
The ruling of the Court of Justice of the European Union in Case C-421/24 of 16 July 2026 confirms an established line of thinking regarding the liability of platforms for content disseminated by users or, as they are now known, content creators.
From this perspective, therefore, the judgement is neither new nor revolutionary, despite having been presented as an innovative watershed between a ‘before’ – when platforms did as they pleased – and an ‘after’ in which this was no longer possible.
The principles set out by the European Court
Before examining the substance of the decision, it is important to bear in mind that the Court of Justice of the European Union is not a court of fourth instance but is concerned solely with certifying the interpretation of regulations and directives, and that it is only national courts that determine civil and criminal liability.
That said, the judgment reaffirms two legal principles.
The first is that if a platform is designed to actively manage content – deciding to whom, when and how to deliver it, and how frequently it should be displayed – this activity falls within the definition of ‘control’ set out in the long-standing Directive 2000/31/EC on electronic commerce. Consequently, the platform in question cannot claim to be a mere passive hosting service that merely makes a service available and, as such, is not liable for the use made of it.
The second legal principle is that an intermediary can no longer invoke technical neutrality once it has acquired sufficiently detailed knowledge of the unlawfulness of the content and fails to take action in accordance with the applicable legislation.
Platforms are not neutral
There can be no doubt that, today, platforms are no longer merely passive curators of user-generated content.
The business model based on user engagement requires that the dissemination of content be optimised to reach, with the greatest possible probability, those interested in consuming it. This involves assessing the merit of what is published, cross-referencing it with both aggregated and individual user profiles, and adapting – in near real time – the way in which content is presented to users logging on at any given moment. In this sense, the European Court has merely reiterated the obvious.
Far more problematic, however, is the assertion that a platform becomes liable for the content it carries if it becomes aware of its unlawfulness by chance or following a report from a third party (the latter being legalised by the creation of ‘trusted flaggers’ — modern delatores — under the EU Digital Services Act).
The critical issues raised by the decision for individual rights
The determination of whether the law has been breached is the sole responsibility of the judiciary and cannot be delegated to a private entity. This applies, in particular, to criminal law, which systematically comes into play when dealing with content that may be defamatory, incite hatred or discrimination, is intended to disrupt public order or economic activities, infringes copyright, or is of an obscene or gruesome nature.
However, whilst there are cases in which the nature of such content is so evident as to leave no room for doubt, there are just as many — some of which have gone to trial, others settled through private agreements, and still others remaining theoretical scenarios — in which the assessment is far from straightforward.
Let us examine a few of these.
Defamation
In a case dating back to 2016 the Court of Cassation held that the operator of a website was liable for keeping false content online, despite having been informed that the reported news was untrue. However, there are just as many cases in which the situation is far more uncertain, as a few examples demonstrate.
The first is the historic (historic for legal experts, of course) court case from 1996 involving Enzo Biagi, who was accused of defamation for having written: ‘Two things are useless: the Italian army and the Pope’s balls. I’d also suggest an addition: The Ombudsman for Publishing’. The investigating judge at the Court of Milan ruled that these statements constituted a critical judgement – one with which one might disagree, but which was not defamatory. How would a platform react today to a statement of that kind?
A few years later, in 2002, the Court of Cassation again ruled on one of the first cases of online defamation, establishing that simply uploading content is not sufficient to constitute the offence, as it is necessary for that content to be actually viewed by someone. The judgement states verbatim: “Posting content online does not in itself constitute an offence against reputation; such an offence will only arise when visitors access the site or when participants in the newsgroup read the message.”
Is it reasonable to expect a platform to decide independently whether content with strong language crosses the threshold of defamation or propaganda, or whether the offence has actually been committed?
Freedom of expression
A critical analysis of controversial content whose dissemination is prohibited — propaganda, conspiracy theories, political choices, ideological views — may reproduce such content to spare the content creator — particularly if they are a journalist — from accusations of distorting the original message.
How can a platform operator determine when the legal threshold prohibiting the dissemination of false news has been exceeded?
Right to be forgotten
An online publication publishes an article on the history of an artist, reproducing newspaper front pages in which, at the start of their career, the artist in question appears with a rather questionable look. The publication receives a request from the artist to remove that front page because the image ‘no longer represents who the artist is today’.
Is it really tenable that a public figure should have the right to have the history of their artistic development erased from the press?
Copyright
The inclusion in a video of content used under a licence granted by a stock media repository is labelled as a copyright infringement even though the source is, quite clearly, lawful. Variations on this theme involve accusing the user of unlawfully performing a melody in the public domain (i.e. not subject to copyright) because there are recordings of the same melody by record labels that commissioned its performance, but which clearly hold no rights to the melody itself.
If not handled correctly — even when one knows how to do so — this situation makes a ‘copyright strike’ highly likely and puts the account’s status at risk.
Dissemination of obscene or gruesome material
The dissemination of images showing the dissection of a human body with internal organs exposed may certainly constitute gruesome content, but if it takes place as part of a pathological anatomy lesson on a medical degree course, it is clear that this is not the issue at hand. In this regard, it is worth noting that even today, content advertising travelling exhibitions of plastinated corpses is still circulating everywhere.
On what grounds should a platform block access to this content?
The dangers of privatised justice
Imposing liability on a platform – or, which amounts to the same thing, an obligation to remove content regardless, in essence, of its actual nature and solely because someone has requested it – means only one thing: that at the slightest hint of doubt, that content will be removed and then, perhaps, there will be a judge in Berlin.
To put it bluntly, this is possible because the path chosen some time ago by the European Union is effectively to privatise the administration of justice, and thus to severely restrict fundamental rights, in particular freedom of expression and the right to a defence.
If content is declared unlawful by a judge, its removal is, of course, required and must be carried out with some speed. If, on the other hand – and as required by the European Digital Services Regulation – a platform is obliged to act systematically to remove content before a judge has ruled on the matter, then when this happens the user will be forced, if they can afford it, to take legal action against the platform. This entails costs that few can afford and therefore, in practice, makes it virtually impossible to have the error rectified and to obtain compensation.
Prevention by preventing defence
It could be argued that this is not, in reality, a curtailment of individual rights but an effective solution to the problem of the dissemination of illegal content, because if people know they cannot defend themselves, they will avoid getting into trouble.
In other words, according to this line of thinking, prevention is achieved by preventing people from exercising their rights and by curtailing the fundamental role of the administration of justice by the state.