Artificial intelligence (AI) is becoming increasingly pervasive in our daily lives, from virtual assistants like Siri and Alexa to autonomous vehicles and medical diagnosis systems. However, the widespread adoption of AI raises questions about trustworthiness, particularly in industries where AI plays a significant role in AI. To ensure trustworthy AI in aviation, the European Union Aviation Safety Agency (EASA) has suggested seven critical aspects in their latest EASA AI Roadmap 2.0: human agency and oversight, technical robustness and safety, privacy and data governance, transparency, diversity, non-discrimination, and fairness, societal and environmental well-being, and accountability.

Human agency and oversight

Human beings must retain ultimate control over the decisions made by AI systems. MROs and lessors must ensure that AI systems operate under human supervision and that humans can intervene in case of a malfunction or unexpected behavior. For instance, when using AI for predictive maintenance, MROs should ensure that technicians review and validate the AI-generated predictions before scheduling maintenance activities.

Technical robustness and safety

AI systems must be resilient and operate safely under different conditions. MROs and lessors must ensure that AI systems are designed to be reliable, secure, and resistant to attacks. For instance, when using AI for aircraft inspection, MROs must ensure that the AI system can detect defects accurately, even when the lighting conditions are poor.

Privacy and data governance

‍AI systems must respect individuals' privacy rights and comply with data protection laws. MROs and lessors must ensure that AI systems collect, store, and use data in a transparent, secure, and ethical manner. For instance, when using AI for predictive maintenance, MROs must ensure that the data collected from aircraft sensors is anonymized and that there are appropriate security measures in place to prevent unauthorized access.

Transparency

‍AI systems must be explainable and enable humans to understand their operation and decisions. MROs and lessors must ensure that AI systems provide clear and concise explanations of their predictions and decisions. For instance, when using AI for fault detection, MROs must ensure that the AI system provides a detailed report of the detected faults and the underlying reasoning.

Diversity, non-discrimination, and fairness

‍AI systems must be designed to avoid bias and discrimination against individuals or groups. MROs and lessors must ensure that AI systems are trained on diverse data sets and that there are no unfair or unintended consequences for specific individuals or groups. For instance, when using AI for predictive maintenance, MROs must ensure that the AI system does not discriminate against aircraft from certain manufacturers or airlines.

Societal and environmental well-being

‍AI systems must be designed and used to promote societal and environmental benefits and to minimize their negative impacts. MROs and lessors must ensure that AI systems contribute to the broader public good and do not harm the environment or society. For instance, when using AI for aircraft scheduling, MROs and lessors must consider the environmental impact of each flight and prioritize more sustainable options.

Accountability

‍AI systems must be accountable for their actions, and their creators must take responsibility for their behavior. MROs and lessors must ensure that AI systems are transparently and ethically designed, monitored, and maintained, and that they can be held accountable for any negative outcomes resulting from their use. For instance, when using AI for predictive maintenance, MROs must ensure that they have appropriate procedures in place to investigate any incidents or accidents related to the AI system's predictions.

‍Conclusion‍

Trustworthy AI is crucial for ensuring that AI systems are designed and used ethically and responsibly. MROs and lessors have many opportunities to use AI to improve their operations, but they must consider each of the seven aspects suggested by EASA.

In embracing Trustworthy AI, we @KeepFlying open doors to a future that harnesses the transformative power of AI while preserving our core values. By cultivating transparency, fairness, ethics, collaboration, and continuous learning, we can shape an AI-driven ecosystem in Aviation that is trusted, accountable, and beneficial to all.

‍

References:

‍