AI disclosure requirements in Hungary: what applies from 2 August 2026

The EU AI Act's transparency obligations are a Union-wide regulation, so the substance is the same in Hungary as everywhere in the EU: from 2 August 2026, AI systems that interact with people must disclose they're AI, synthetic media must carry machine-readable marking, and deepfakes and certain AI-generated public-interest text must be labeled. What differs country by country is who enforces it, in what language your users encounter it, and how quickly complaints turn into questions — which is what this page covers for Hungary.

Enforcement in Hungary

Hungary's Act LXXV of 2025 has been in force since 1 December 2025, making it one of the earlier national AI Act implementing laws in the EU — though two provisions (Section 3(2) and Section 10) are held back until 2 August 2026, aligning with Article 50 itself. The National Accreditation Authority (NAH) is the notifying authority, accrediting the conformity-assessment bodies that certify high-risk AI systems. The Minister responsible for Enterprise Development, within the Ministry of National Economy, is the primary AI Market Surveillance Authority and single point of contact — investigating non-compliance, ordering corrective measures, and overseeing the national regulatory sandbox, which is due to be fully active by August 2026. A new Hungarian Artificial Intelligence Council provides strategic guidance and cross-government coordination. Article 50 applies from 2 August 2026 with the EU's penalty ceiling of €15 million or 3% of worldwide turnover.

The four duties, as they show up on your site

• Article 50(1) — chatbot disclosure. Any AI system interacting with visitors must make that clear no later than the first interaction, in a clear and distinguishable way. A line in your terms doesn't satisfy it.
• Article 50(2) — machine-readable marking. AI-generated audio, images, video, and text must be marked in a machine-readable, detectable format. For generative systems already on the market before 2 Aug 2026, this specific duty has a transition until 2 Dec 2026 — the rest does not.
• Article 50(3) — emotion recognition & biometric categorisation. People exposed must be informed the system is operating.
• Article 50(4) — deepfake and AI-text labels. Visible disclosure for deepfakes, and for AI-generated or manipulated text published to inform the public on matters of public interest.

Hungary-specific practicalities

• Hungary's supervision runs through a government ministry (Enterprise Development) rather than an independent regulator — expect a more centrally-directed enforcement posture than in member states using arm's-length agencies
• The law itself has been live since December 2025, so the institutional machinery (single point of contact, sandbox, Council) has had a runway to mature well before Article 50's own August 2026 start date
• The regulatory sandbox reaching full activation by August 2026 makes it a realistic pre-launch option for testing a disclosure or labelling approach before EU-wide enforcement begins
• Keep the Hungarian-language first-interaction screenshot in your evidence file, dated, alongside your widget settings export

Disclosure language should match your user-facing language. For a site serving Hungary in Hungarian, a first-message line such as "Ön egy mesterséges intelligencia (AI) asszisztenssel beszélget" (you're chatting with an AI assistant) is the pattern to aim for — at first interaction, not in the általános szerződési feltételek.

Common questions