On 27 October 2026 the European Commission proposed a new mandate for Frontex. The most consequential part is not the budget. It is a small mechanism: an Online Monitoring and Response Centre inside a border agency, paired with an intention to seek trusted-flagger status under the Digital Services Act (DSA).
The case for the proposal
The strongest argument for the package is coherence. EU return policy has long been fragmented across 27 national systems, and the Commission says it wants a dedicated Return Coordination Centre and a standing corps of 30,000 officers. Smuggling networks also advertise routes and services openly on social media. A border agency that cannot read public posts is working with one hand tied. The Commission describes the new centre as a way "to detect dangerous trends early, improve information exchange and enable quick intervention." Monitoring publicly accessible content is also not the same as reading private messages.
That is a legitimate objective. The question is whether the design fits it.
What the proposal actually does
According to the Commission's news item, the Online Monitoring and Response Centre would monitor publicly accessible social media content. netzpolitik.org reports that Article 33 of the draft would cover online content that facilitates, promotes or gives guidance on unauthorised border crossings. The same report says Commission Vice-President Henna Virkkunen indicated Frontex will apply for trusted-flagger status and expects approval because of its expertise. MLex summarises the effect: platforms would receive Frontex notices about illegal content linked to border security risks, and a designation would give those notices priority.
Advocacy group Protect Not Surveil, a partisan source, adds that the reform gives Frontex powers to research, test and procure AI-enabled tools, drones and sensors. I could not find that procurement claim in other reporting on the 27 October text. The group also cites about €12 billion for 2028–2034, a figure that comes from the Commission's separate 2028–2034 budget proposal rather than from the 27 October regulation itself. The Commission pages I could access do not state a budget figure, and other reports put the request at €11.8 billion. Treat the number as a reported proposal, not an adopted allocation.
Why trusted-flagger status is the wrong fit
Trusted flagging under Article 22 of the DSA (Regulation (EU) 2022/2065) was designed for specialised bodies that bring expertise to notice-and-action. Platforms must process their notices with priority. The design assumes a flagger whose incentives are accuracy and whose status can be withdrawn if it errs. Typical examples are hotlines on child abuse material or consumer protection groups.
A border enforcement agency is a different animal for three reasons.
- Incentives. Frontex's institutional goal is to reduce arrivals. An agency measured on that outcome has every reason to flag broadly. Over-flagging costs it nothing.
- Category creep. The draft targets content that "facilitates" or "promotes" crossings. That is far vaguer than content that is illegal under a defined national or EU offence. Much of it will be journalism, NGO search-and-rescue communication, diaspora advice and legal-aid information, all of which is lawful speech.
- Priority treatment. A priority notice from a state agency changes platform behaviour. Platforms that fear regulatory friction tend to comply first and ask questions later. The DSA's own logic is that fast-track channels need strong accountability, which an agency with a large enforcement mandate does not obviously supply.
The Commission's line, as reported, is that the DSA enforcement framework remains fully respected. That is true on paper, since the designation would be made by a national Digital Services Coordinator. But those coordinators are unevenly resourced, and a designation sought by a Commission-backed agency, with a Commission Vice-President saying she expects it to succeed, is not a neutral test of the criteria.
Surveillance without a defined evidentiary standard
The monitoring centre also raises a data protection problem. Feeding aggregated social media signals into Frontex's risk analysis, as critics describe it, means processing personal data of people who are not suspected of any offence. The EU has spent a decade building purpose limitation and necessity tests into law precisely to prevent that kind of open-ended collection. A proportionate design would define the offences in scope, require human review of every notice before it goes to a platform, publish notice volumes and error rates, and give affected users a route to challenge a removal.
None of that is visible in the summaries so far. The full text, available as a Commission proposal document, deserves close reading by Parliament and the Council before anyone treats the safeguards as settled.
A proportionate alternative
Nothing here requires Frontex to become a flagger. Three narrower options would deliver most of the operational value.
- Let Frontex share findings about smuggling advertisements with Europol and national law enforcement, who can use existing legal orders under national criminal law, with judicial oversight.
- If trusted-flagger status is pursued, confine it to content that is unlawful under a specific criminal offence, not to broad categories of migration-related speech.
- Require transparency reporting on notices and outcomes, with independent audit by the data protection authorities.
A pro-innovation, pro-speech approach does not mean ignoring smuggling. It means refusing to build a standing channel from a border agency to platform moderation queues before the accountability rules exist. Once such a channel is in place, it is rarely narrowed.
What to watch
The proposal is at an early stage. Parliament and the Council must now decide how far Frontex's powers should expand. The amendments that matter most are not on budget or headcount. They are the definition of in-scope content in Article 33, a bar on designation as a trusted flagger or a strict condition on it, and a ban on using the monitoring centre's outputs for individual-level profiling.