Jeppesen ForeFlight, the Colorado-based aviation technology company, has introduced ForeFlight Airflow, an artificial intelligence engine designed specifically for aviation safety that aims to reduce flight plan data errors without replacing human pilots.
Unlike general-purpose generative AI, ForeFlight Airflow combines AI with flight-planning data, operational records, and aviation regulations. The system evaluates operational context before generating recommendations, drawing simultaneously from multiple sources including weather data, airport restrictions, fuel requirements, crew requirements, and airspace rules.
“Artificial intelligence is not enough for this industry, we need aviation intelligence: the assurance that the right data, right context, and right reasoning are applied every time,” said Brad Surak, CEO of Jeppesen ForeFlight.
How it works
The system is model-agnostic, meaning it can work with multiple AI backends, internal models, third-party platforms, or user-chosen ones. Its outputs are explainable: the system shows the data, constraints, and alternatives it evaluated, rather than delivering opaque recommendations. The technology is integrated into existing aviation safety-management practices, including hazard identification, safety trend analysis, predictive risk modeling, and operational data integration.
First product: AI Connector for ChatGPT
The initial product built on ForeFlight Airflow is the ForeFlight AI Connector, a Model Context Protocol (MCP) server that connects ForeFlight Mobile to a customer’s existing OpenAI ChatGPT environment. Pilots can ask natural-language questions to retrieve route options, fuel stops, and weather details. Future expansion to Google’s Gemini and Anthropic’s Claude is planned, allowing users to choose their AI experience layer.
Why aviation needs its own AI
Aviation decisions depend on interconnected factors where errors carry fatal consequences. Conventional LLMs hallucinate and provide answers without sourcing, unacceptable in a safety-critical setting. ForeFlight Airflow addresses this by filtering outputs through aviation-specific governance protocols, making reasoning visible, and ensuring recommendations are reviewed through established safety principles.
The system is designed to support human judgment, not replace it. Critical decisions remain in human hands while the AI processes large volumes of operational information.
What comes next
The initial focus is general aviation, with future offerings planned for commercial, business, and military aviation. Regulatory acceptance and real-world testing will determine deployment timelines in safety-critical environments. If the approach proves successful, it could establish a model for how AI is introduced into any industry where accuracy and accountability matter more than generating a quick answer.