A survey by European aviation regulator EASA found that aviation experts felt “cautiously optimistic” about the use of AI in the industry, and their biggest concerns are privacy, safety, surveillance and skill removal.
A survey of 231 aviation experts was conducted in January 2024 to explore eight virtual AI application scenarios and ask questions about their level of comfort, trust and acceptance.
On average, experts rated the acceptance of AI at 4.4 out of 7, reflecting careful optimism about its potential and raising concerns about its risk. Approximately two-thirds of respondents expressed reservations and rejected at least one AI scenario.
Respondents were most concerned about AI performance, data protection and privacy, accountability, and potential safety limitations.
The survey also found that the majority highlighted the need for robust regulation and supervision by AI's EASA and the National Aviation Authority. Furthermore, participants were concerned that human knowledge and capabilities could be reduced if AI took over partially or completely.
The full report is available on the EASA website. Ethics of AI in Aviation – Aviation Expert Survey Results 2024/2025 | Ether.
Commenting on the research by Guillaume Soudain, AI Program Manager at Easa, he said, “AI offers a great opportunity to improve aviation safety and efficiency, but trust is important.
“This study highlights the importance of a balanced regulatory framework, which ensures the highest level of safety for citizens while promoting innovation and competitiveness in the European aviation sector.”
AI DAYS Event
At the EASA AI Days event in Cologne where the investigation was presented and discussed, Christine Berg, director of aviation safety at the European Commission, said:
“Aviation is, by definition, critical of safety. This requires a system that is not only intelligent, but also explainable, reliable, and proven.”
During the AI Days event, EASA representatives shared research on AI assurance, human factors, and ethics-based assessments. The series of workshops gathered feedback from industry, authorities, research centers and universities to inform EASA's “AI Roadmap.”
Consultations from the FAA and Eurocae provided a broader perspective on how AI aviation development can be applied and strengthened the joint spirit between these organizations and the EASA.
The second day of the meeting focused on AI research. Boeing shared a survey of automated tax AI-based systems, and Luftansa Group spoke about surveys of large-scale language model (LLM)-based troubleshooting assistants.
The day also included a discussion panel on human surveillance with Level 3 AI and AI-based operational tools within approved aviation organizations.
Three projects undertaken under the auspices of SESAR-JU have been announced and discussed: Jarvis (a rather intelligent system), Darwin (a digital assistant to reduce workloads and collaboration), Honealwell and Synthair (improvement of AT automation and simulation with AI-based universal model of Syntetic Data Generation).
Find recordings and presentations for AI Days Conference here.
