Toward a Decision-Assurance Layer for AI-Assisted Flight Planning in Air Traffic Management
This arXiv paper proposes the AI Trust and Assurance Layer (ATAL), a model-agnostic decision-assurance architecture that evaluates whether AI-generated flight-planning outputs are reliable enough for operational use by combining measures of semantic stability under prompt variation, operational consistency of structured outputs, and normative constraint validation, and mapping those signals to a Decision Readiness Level (DRL). An ATM-inspired experimental study demonstrates how unsafe, inconsistent, or misleading outputs can be identified before they affect flight-plan validation or execution, and the authors suggest the framework is transferable to other safety-critical, human-supervised domains.