Artificial intelligence (AI) is quickly changing the aviation industry. To safely bring AI into the skies, the European Union Aviation Safety Agency (EASA) created a specific plan called the EASA AI roadmap. From helping mechanics predict when an airplane needs maintenance to helping air traffic controllers manage flights, AI tools are becoming very useful. However, aviation is a safety-critical domain where human decisions and systems must be highly reliable. Because of this, we cannot treat AI as just another digital tool.

Here is a simple breakdown of what the EASA AI roadmap is all about and why it is critical for modern aviation professionals.

What is the EASA AI Roadmap?

The EASA Artificial Intelligence Roadmap 2.0 provides the strategic foundation for using AI in aviation. It describes the new opportunities AI creates and identifies the challenges that must be addressed before AI can be widely and safely adopted. Most importantly, the EASA AI roadmap introduces the idea of “trustworthy AI” for aviation. Trustworthy AI means that we don’t just care if the technology works; we need evidence that it behaves as intended, that humans can interact with it effectively, and that safety risks are managed appropriately.

The Four Pillars of the EASA Framework

To make AI safe, the EASA framework uses a structured approach built on four main building blocks. These pillars are the core focus of EASA’s regulatory strategy and are further detailed in EASA’s Concept Papers and rulemaking tasks:

• Trustworthiness Analysis: Before using AI, we must clearly describe what it is intended to do, where it will operate, and who will interact with it. This step also involves an ethics-based assessment to ensure the system allows for human oversight and is transparent.
• AI Assurance: This block provides the structured evidence needed to show that the AI system satisfies its requirements and is suitable for aviation use. It focuses heavily on data quality, model training, and verifying that the AI behaves correctly in its intended operational environment.
• Human-Centred Design Considerations: Aviation safety depends on people and technology working well together. This block focuses on “Human-AI Teaming,” ensuring that pilots, engineers, or air traffic controllers can understand the AI’s outputs, supervise it, and use it safely.
• AI Safety Risk Mitigation: This final block focuses on managing any safety risks that remain after the first three steps are addressed. It involves putting safeguards and operational procedures in place, which is especially important for advanced automation where direct human oversight might be reduced.

Why Aviation Needs Specific AI Rules

Unlike traditional software that runs on strict rules, many AI applications learn their behavior from data. This means an AI’s behavior can be context-dependent and difficult to predict in every possible situation. EASA’s guidance helps classify AI based on its role. For example, a Level 1 AI might simply assist a human with decision-making, while a Level 3 AI might perform advanced automated actions that require special safeguards.

Furthermore, organizations need to be ready for AI. They need the right processes to manage data, monitor the AI’s performance once it is in service, and handle any unexpected behaviors.

Master Trustworthy AI in Aviation

Whether you are an AI developer, a pilot, an engineer, a safety manager, or a regulator, understanding the EASA AI roadmap is becoming an essential skill. You do not need any previous experience in artificial intelligence or coding to grasp these concepts.

Trustworthy AI in aviation is the future of the industry. To gain a foundational understanding of how to safely introduce and supervise AI in accordance with emerging European expectations, enroll in our comprehensive online course today: Trustworthy AI in Aviation: Understanding the EASA Framework.

Key EASA Documents & References

If you want to dig deeper into the official regulatory texts that shape this framework, here are the actual documents guiding trustworthy AI:

• EASA Artificial Intelligence Roadmap 2.0: The strategic document that establishes the foundation for a human-centric approach to AI in aviation and introduces the core building blocks.
• EASA Concept Paper: Guidance for Artificial Intelligence Applications (Proposed Issue 03): The latest technical guidance expanding the scope to advanced automation (Level 3) and addressing additional AI techniques like reinforcement learning and symbolic AI.
• Regulation (EU) 2024/1689 (EU AI Act): The harmonized European legislative framework for AI, which identifies specific aviation domains and requirements under Article 108.
• Notice of Proposed Amendment (NPA) 2025-07: EASA’s formal rulemaking proposal providing detailed specifications (DS.AI), acceptable means of compliance, and guidance material for AI trustworthiness.