Part I — Situation overview

On the morning of 17 July 2026 it became public, from the data set released by the Ministry of Justice in response to a freedom-of-information request by 24.hu: over the 16 years of the previous governments, the justice ministers authorised national-security secret surveillance in a total of 18,906 cases, and said no in a mere 50. The most authorisations went to the Constitution Protection Office (30 refusals out of 10,403 requests); of the Counter-Terrorism Centre’s 3,740 requests three were refused, of the Information Office’s 3,320 requests thirteen, and every request of the Military National Security Service went through. The number of requests grew continuously, peaking in 2021 — the year of the Pegasus scandal — with 1,469 authorisations. “The authorisation rate is practically one hundred per cent, which means the minister exercises no real control” — was the assessment of the data by Ádám Remport, researcher at the TASZ (Hungarian Civil Liberties Union, a rights-defence organisation).

The numbers make a long-known structural fault visible in quantitative terms for the first time. Hungarian regulation — the national security act, and in the case of the Counter-Terrorism Centre the much-debated section 7/E of the police act — ties national-security secret information gathering, exceptionally in an EU comparison, to the executive, to ministerial authorisation, without judicial control. The ECtHR (European Court of Human Rights, the Strasbourg human rights court) condemned Hungary for this in 2016 in Szabó and Vissy v. Hungary, and the 2021 Pegasus affair — in which the phones of journalists and civilians were infected with spyware — demonstrated the potential for abuse. It is natural that counter-intelligence works on a broader front than law enforcement; what is not natural is that the approver practically never says no.

MIAK’s reading: the 19,000-strong data set is not about every surveillance having been unlawful — it is about there being no point in the system where anyone could have checked this independently. Ministerial authorisation worked not as a filter but as a rubber stamp, and this is one of the most concrete, internationally standard-backed terrains for restoring checks and balances.

Part II — Literature foundation

Before turning to MIAK’s proposals, it is worth fixing the interpretive frame. Shoshana Zuboff (professor emerita of Harvard Business School), in The Age of Surveillance Capitalism, describes surveillance as a technique of power: whoever commands behavioural data is capable not only of knowing but of shaping people’s behaviour — unchecked surveillance capacity is thus a factor of power in itself, regardless of whether it is currently being abused. According to Spin Dictators by Sergei Guriev and Daniel Treisman (an economist of Russian origin and an American political scientist, respectively), modern autocracies typically keep society in hand not with open violence but with informational instruments — among them advanced surveillance technologies —, which is why the institutional limitation of surveillance is a foundational question of democratic resilience. And the European Convention on Human Rights (Article 8) provides the legal yardstick: interference with private life is permitted only when regulated by law and necessary in a democratic society — this test is the measuring rod of the Hungarian reform. The detailed literature treatment — by author, with quotations — can be found in the 6.4 Literature in detail section.

Part III — MIAK’s concrete proposal

MIAK proposes three measurable measures.

3.1 Judicial authorisation for national-security surveillance (legislative amendment within 12 months)

The National Assembly should amend the national security act so that national-security secret information gathering is authorised by a judge — or, in exceptional cases that brook no delay, by an independent quasi-judicial body subject to ex-post judicial confirmation —, in line with the ECtHR’s Szabó and Vissy judgment. Surveillance for law-enforcement purposes is authorised by an investigating judge even today; there is no principled reason why the national-security branch should stand under less strict control, when there the surveilled person never learns of the proceeding. This is the direct application of the A6 checks-and-balances programme point, and it creates the institutional guarantee of the Convention test (see 6.4.3).

3.2 An annual public surveillance statistic (first report in spring 2027)

The government should publish annually the number of national-security authorisations, broken down by service and by legal basis — as has now happened following a one-off data request, only made regular. The individual case is a secret; the aggregate number is not. The public time series is the only instrument by which the public and parliament can see if the number of authorisations spikes without justification — from the present data set, for example, it can be read that the peak year was precisely the year of the Pegasus scandal. The statistic would supply the surveillance indicator of the A9 Spin-dictatorship prevention index, and would also feed into the annual database of the I10 constitutional stress test, following the logic of the Guriev–Treisman frame (see 6.4.2): against the “soft” instruments of repression the defence is measurability.

3.3 Ex-post notification on the German model (by the end of 2027)

Whoever was surveilled without the surveillance leading to any measure must be notified afterwards — after the national-security interest has lapsed, with a statutory deadline and set of exceptions —, as German constitutional practice prescribes. Ex-post notification is what enables the person concerned to turn to a court, that is, it closes the remedial loop whose absence the ECtHR objected to. In the retrospective disclosure — the handling of the 19,000-strong stock that has now come to light — the same principle applies: not public listing, but an individual, rule-of-law notification procedure.

The common principle of the three proposals: security and liberty are not a zero-sum game — surveillance without control weakens security too in the long run, because it undermines the services’ credibility and their capacity for international cooperation as well. Zuboff’s thesis (see 6.4.1) holds for state surveillance too: unchecked surveillance capacity is power in itself, and so its oversight is not a question of trust but of structure.

Part IV — Expected effects and risks

Dimension Expected effect Risk
Rule of law Judicial authorisation closes the ECtHR condemnation open since 2016; the services’ legitimacy strengthens The reform remains a facade if no substantive information reaches the judge — the “rubber stamp” merely changes hands
National security Legitimate surveillance becomes more credible; the trust of international partner services grows Too rigid a procedure may slow the genuinely urgent cases — the exception that brooks no delay requires careful regulation
Society The public statistic and ex-post notification restore trust The retrospective disclosure can also be framed as political score-settling — impartial handling of the individual cases is critical

The main dilemma stretches between speed and guarantee: counter-intelligence may genuinely need a decision within hours, so a well-regulated urgency procedure must stand beside judicial authorisation — with ex-post confirmation, not an uncontrolled loophole. The proposal works if the judicial side receives capacity and a right of insight as well, not merely a signing task.

Part V — Measurability and summary

5.1 What is worth tracking? (proposed KPIs)

MIAK proposes tracking four key performance indicators (KPIs):

  • Legislative amendment: submission of the proposal introducing judicial (or quasi-judicial) authorisation within 12 months, adoption by the end of 2027.
  • Refusal rate: the refusal rate of authorisation requests under independent control — instead of the current 0.26%, a substantive filtering of several per cent would signal real control.
  • Public statistic: appearance of the first annual surveillance report by spring 2027, then annually.
  • ECtHR compliance: closure of the execution of the Szabó and Vissy judgment before the Committee of Ministers of the Council of Europe.

5.2 Summary

MIAK asks the National Assembly and the government to treat the 19,000-strong data set not as a scandal but as a blueprint: the numbers show precisely where control is missing. The three steps — judicial authorisation, an annual public statistic, ex-post notification — together close the gap that the ECtHR has been signalling for a decade.

Two MIAK foundational values are at stake in the case. Accountability is to be taken literally here: secret instruments too need an answerable decision-maker — judicial precisely because the minister is not independent of the government. Transparency, in turn, applies at the level of aggregate data: the individual case of surveillance may be a secret, the workings of the system may not. The yardstick here too is single: the same control applies to every government — including the present one.

Part VI — Justifications and further sources

6.1 The press framing by spectrum

The topic was dominated by the public-affairs and left-liberal band, which is no surprise: the source is 24.hu’s own data request. 24.hu processed the data set in a subscription analysis, with three experts involved, focusing on the system-level absence of control (the full article was not publicly downloadable). Telex presented the numbers and the expert assessments, highlighting the Constitution Protection Office’s outsized caseload and the interpretation of Transparency International and the rights-defence experts. HVG put the thesis “there was practically no ministerial control” in its headline, and tied the Pegasus thread and the Strasbourg condemnations to the data.

The pro-government-conservative band (Magyar Nemzet, Mandiner) expressly did not rank the topic among its highlighted themes that day — the front pages were dominated by the debate on the constitutional transition and the government-critical framing of the Szijjártó affair. The economic band (Portfolio) likewise did not treat it. This framing asymmetry is itself telling: the surveillance data set is comfortable for neither political side — for the previous one because of its size, for the new one because of its continuity —, which in MIAK’s assessment is precisely why it is suited to serve as the starting point of a party-neutral institutional reform.

6.2 Facts and data

Service Authorisation requests (16 years) Refusals
Constitution Protection Office 10,403 30
Counter-Terrorism Centre 3,740 3
Information Office 3,320 13
Military National Security Service 1,370 0
National Security Special Service 122 n. a.
Total (as reported by 24.hu) 18,906 50

The source of the per-service numbers is the press summary; their sum (18,955) differs slightly from the 18,906 total reported by 24.hu — the source does not explain the discrepancy.

  • Source: the Ministry of Justice’s response to 24.hu’s freedom-of-information request (17 July 2026).
  • The annual number of authorisations rose continuously; peak: 2021, with 1,469 authorisations.
  • ECtHR, Szabó and Vissy v. Hungary (37138/14), 2016: the Hungarian regulation does not provide sufficient guarantees against abuse.
  • The final report of the European Parliament’s Pegasus committee of inquiry (2023) treated Hungarian surveillance practice in a dedicated chapter.

6.3 Policy dimensions

  • Justice (programme points) — extending judicial control to national-security authorisation, constitutional stress test (I10);
  • Transparency and anti-corruption policy (programme points) — restoring checks and balances (A6), measuring democratic resilience (A9);
  • Defence (background) — the balance between the democratic control of the national-security services and information protection (HV11).

6.4 Literature in detail

6.4.1 Shoshana Zuboff: The Age of Surveillance Capitalism

Zuboff’s work analyses surveillance not as a technical but as a power phenomenon: the mass collection of behavioural data creates a new kind of power — she calls it “instrumentarian power” — that is capable not only of knowing people’s behaviour but of shaping it according to others’ ends. Although Zuboff writes primarily about market data collection, her thesis holds all the more for state surveillance: where surveillance capacity is unchecked, so is power, regardless of the intentions of whoever wields it at any given time. In this frame the Hungarian data set is to be read thus: the question is not how many abuses lay behind the 18,906 authorisations — but that, because of the system’s structure, no one can know.

📖 Source: Shoshana Zuboff: The Age of Surveillance Capitalism

6.4.2 Sergei Guriev and Daniel Treisman: Spin Dictators

According to Guriev and Treisman’s thesis, modern autocracies are built on informational control instead of open violence: alongside the manipulation of the media, it is advanced surveillance technologies that make repression possible to render “invisible”. The authors mention the Hungarian system too among the cases examined. The book’s practical lesson for the reform is that defence against surveillance capacity is not a one-off legislative amendment but a matter of continuous measurement: what must be made visible is how much and for what the instruments are used — precisely what MIAK’s annual public statistic proposal and the index-like tracking of democratic resilience aim at.

📖 Source: Sergei Guriev–Daniel Treisman: Spin Dictators

6.4.3 European Convention on Human Rights (Article 8)

Under Article 8 of the Convention, “everyone has the right to respect for his private and family life, his home and his correspondence”, and a public authority may interfere with this right only where it is “in accordance with the law and is necessary in a democratic society” — among other grounds, in the interests of national security. It was from this test that the Strasbourg court derived the requirements of the Szabó and Vissy judgment: the body authorising surveillance must be independent of the executive, and the person concerned must be given an ex-post possibility of remedy. The Hungarian 19,000-strong data set documents the absence of both elements of the test — the reform is thus not a matter of political taste but of Convention obligation.

📖 Source: European Convention on Human Rights (Article 8)

6.5 International comparison

The German model is the most frequently cited example: national-security surveillance is authorised by the so-called G10 Commission — a judicial-type body elected by parliament —, those concerned must as a rule be notified after the surveillance has ended, and the constitutional court has narrowed the services’ room for manoeuvre in several judgments. In the Netherlands, since 2018 an independent legal commission (TIB) approves authorisations in advance, with two-tier control. In the United Kingdom, the 2016 act introduced a “double lock” system: the ministerial warrant is also approved by a judicial commissioner. The common element everywhere: nowhere does the executive authorise on its own — some independent, judicial-type actor is always part of the authorisation chain. The Hungarian system is the exception in this field, and the ECtHR said so as early as 2016.

Transparency and anti-corruption policy

  • A6 — Strengthening checks and balances
  • A9 — Spin-dictatorship prevention index

Justice

  • I10 — Constitutional “stress test”

Defence

  • HV11 — Strategic communication and information protection

Proposed new programme point: Judicial authorisation and an annual public statistic of national-security surveillance — for the Justice area.

6.7 List of sources

Press sources (MIAK press monitor, 17 July 2026 — topic 2):

Knowledge-base references (literature):

  • 📖 Shoshana Zuboff: The Age of Surveillance Capitalism
  • 📖 Sergei Guriev–Daniel Treisman: Spin Dictators
  • 📖 European Convention on Human Rights (Article 8)

MIAK internal materials:

  • MIAK policy area: Transparency and anti-corruption policy (programme points; programme point ID: A6, A9)
  • MIAK policy area: Justice (programme points; programme point ID: I10)
  • MIAK policy area: Defence (programme points; programme point ID: HV11)
  • MIAK press monitor, 17 July 2026 — topic 2, score: 90/100

Additional public data sources:

  • ECtHR, Szabó and Vissy v. Hungary (37138/14), 2016
  • Venice Commission, opinion CDL-AD(2016)012
  • European Parliament PEGA committee final report (2023)

Generation metadata