Skip to content
Cardan-AI
Back to news
AI Regulation10 July 2026

Cybersecurity and AI: EU Commission action plan targets energy, transport and industry

On 7 July 2026 the European Commission unveiled an action plan linking cybersecurity and AI: an advanced-model evaluation capacity by 2027, an ENISA reference framework for structured access to AI in cybersecurity, and a secure testing platform for critical-sector operators (energy, finance, health, transport, public administration).

On 7 July 2026, the European Commission unveiled an action plan explicitly linking cybersecurity and artificial intelligence. The goal is twofold: govern the risks advanced AI models pose to digital security while accelerating their defensive use to strengthen critical infrastructure resilience. A European capacity to evaluate models before market entry is expected to be operational by 2027, with an explicit focus on cyber vulnerabilities. In parallel, the Commission and ENISA are building a reference framework giving public and private organisations structured, transparent access to advanced AI capabilities for cybersecurity purposes.

The key point for aerospace, O&G and energy: ENISA and the Joint Research Centre will open a secure testing platform letting critical-sector operators — energy, finance, health, transport, public administration — trial AI applications in simulated environments without exposing production systems. This is a rare entry point: testing high-risk AI use cases (intrusion detection on SCADA systems, predictive maintenance on sensitive infrastructure) in an EU-backed sandbox rather than on one's own operational assets.

Cardan-AI's take: this plan reflects a structural shift. AI is no longer seen purely as a cyber risk surface but as a genuine defensive tool — the Commission explicitly recommends leveraging open-source models to detect and fix vulnerabilities faster, under a security-by-design logic. For heavily regulated, exposed industries such as energy, aerospace and defense, this is a strong signal: tomorrow's cyber compliance will natively embed an AI component, and operators who anticipate this convergence gain both a regulatory and operational head start.

Cardan-AI recommendation: from the plan's entry into force on 2 August 2026, map your existing AI-cybersecurity use cases and identify which could benefit from the future ENISA/JRC testing platform — an opportunity to validate sensitive deployments with EU framework support rather than facing the regulator alone.

Analysis by

Cardan-AI Intelligence

Our research and analysis unit, dedicated to applied AI for business, industry and regulatory compliance.

Let's talk about your next competitive edge

Thirty minutes to identify the two or three use cases in your operations that pay for themselves within the first year.