15/07/2026
The Commission presents the EU Action Plan on Cybersecurity and Artificial Intelligence
The European Commission has presented an Action Plan to provide a structured response to address the risks and capitalise on the opportunities presented by advanced artificial intelligence (AI) models for cybersecurity.
New advanced AI models are redefining cybersecurity. AI can be misused to detect vulnerabilities, automate attacks and increase the scale and speed of cyber incidents at an unprecedented rate.
Building on the EU’s single legal framework for AI and cybersecurity, the Action Plan will bring together Member States, industry and EU organisations to strengthen the cybersecurity of our digital environment in the face of the vulnerabilities posed by advanced AI.
Assessment of AI models
Effective security requires a thorough understanding of how new technologies can be used correctly, misused or exploited. Under the Artificial Intelligence Regulation, advanced AI models must be assessed, and mitigation measures must be carefully weighed up before such models are placed on the EU market.
In order to foster specialist expertise within the EU, the Commission will launch a specific call for proposals to establish an EU assessment capacity covering cybersecurity, which is expected to be operational by 2027. This new capacity will support the regulatory role of the AI Office by strengthening the independent assessment of AI capabilities and risks on a global scale.
Access to advanced AI models
Europe also needs clear and transparent conditions for accessing the most advanced AI systems.
The Commission will work with the European Union Agency for Cybersecurity (ENISA) to define a European plan for structured access to advanced AI capabilities for cybersecurity. These guidelines will enable relevant European organisations, both public and private, to access advanced AI models.
Testing AI for cybersecurity
ENISA and the Commission’s Joint Research Centre will create a secure platform for testing AI applied to cybersecurity, which will include the use of simulated environments. This will provide technical expertise on the secure use of AI to operators in critical sectors, such as finance, energy, healthcare, transport and public administration.
Strengthening the EU’s cybersecurity and addressing vulnerabilities
The EU must protect its critical infrastructure against vulnerabilities arising from the potential misuse of these technologies.
As set out in EU cybersecurity standards, organisations must strengthen their cyber hygiene practices, risk management measures and ‘security by design’ principles.
Organisations should start using the AI capabilities already available, including through open-source models, to identify and address vulnerabilities more quickly, as well as to prevent and respond to cyber-attacks.
To assist organisations in this transition, ENISA will support and facilitate partnerships between public authorities, businesses and open-source communities within the cyber ecosystem. This will include guidance, recommendations and best practices, as well as a campaign to protect critical open-source software.
Expanding European AI capabilities for cybersecurity
To stimulate the growth of the European market, the Commission will launch the ‘EU Grand Challenge on AI for Cybersecurity’. This competition will bring together businesses, researchers and organisations with the aim of developing artificial intelligence solutions for cybersecurity.
The EU must continue to invest in the development of its own advanced sovereign AI capabilities, drawing on the infrastructure provided by AI factories and future gigafactories. In this context, the future capital fund for the European technology sector, announced in the Technological Sovereignty Package, could attract private investment to expand home-grown AI capabilities.
Background
The EU has an appropriate legal framework to address cybersecurity in the era of emerging technologies, such as artificial intelligence. The AI Regulation requires the risks arising from AI models to be assessed and mitigated, whilst the Code of Practice on General-Purpose AI sets out these requirements in greater detail and facilitates compliance by providers of advanced models. These provisions will come into force on 2 August 2026.
The Cyber Resilience Regulation (applicable from late 2027) requires hardware and software products to incorporate security by design. In addition, the Directive on Network and Information Systems (NIS 2) aims to enhance security in critical sectors such as transport and energy, alongside the Regulation on Digital Operational Resilience in the financial sector. The Cyber Solidarity Regulation strengthens the EU’s capabilities to detect, prepare for and respond to serious, large-scale cybersecurity threats and attacks.
