Legal blog

New Cybersecurity Rules from 2026

2026-01-09 | IT jog, IT Law

The legislator has adopted Act CXXXV of 2025 amending the relevant legislation, which introduces fundamental changes to the Hungarian cybersecurity regulatory framework.
The purpose of the amendment is twofold: on the one hand, to implement the EU Cyber Resilience Act (CRA) at national level, and on the other hand, to clarify and extend the cybersecurity rules already in force in Hungary. The changes will enter into force gradually from 2026 and will affect a wide range of organisations that have previously not been subject to cybersecurity obligations, or only to a limited extent.

Continue

Search

Last blogs

What Is the EUDI Wallet, and What Can It Be Used For?

2026-10-21

The EUDI Wallet (European Digital Identity Wallet) is a digital solution that enables individuals to electronically prove their identity and present certain personal data and entitlements. Its legal basis is established by Regulation (EU) 2024/1183, which amends the eIDAS Regulation.

In practice, this could mean presenting a digital document proving an educational qualification, verifying one's identity when accessing an online service, or making administrative procedures in another EU Member State easier. The aim is to enable users to verify the information required of them securely and conveniently, with less administrative effort.

The Legal Price of ChatGPT: Copyright and GDPR Mazes Behind Artificial Intelligence

2026-09-11

The revolution of artificial intelligence has forced not only the technology sector but also legislators to take action. Although ChatGPT and similar large language models may appear to users to be simple chat programs, an astonishing amount of data processing takes place in the background. The European Union’s legal framework has highlighted that these models raise serious copyright and data protection challenges, which cause significant headaches for developers.

Know Your Agent – Who Is Liable When AI Acts on Our Behalf?

2026-09-10

The next stage in the development of AI is no longer simply about artificial intelligence answering questions or making recommendations. So-called AI agents are increasingly capable of acting autonomously on behalf of users: initiating purchases, booking appointments, carrying out financial transactions, or even managing processes across multiple digital systems.

This, however, raises a new question, primarily of a legal nature: how can it be determined whether an AI agent was actually authorised to perform a particular action, and who is liable if it exceeds the limits of its authority?