The present research was designed to provide evidence into why and when Unidentified Aerial Phenomena (UAPs) occur and pose a threat to aviation safety. Specifically, the goal was to understand how causal illusions interact with perceptual biases with and without irrational reasoning. A total of 408 airline pilots participated in an experiment using Airbus® aircraft TCAS/ROSE simulators. Analyses were conducted using structural equation modeling (SEM), controlling for the effects of fatigue and flight hours. Results indicated that causal illusions were 82.4% predictive of UAP sightings only when magical inference was present. Our experimental evidence shows that UAPs may be explained as cognitive biases and would pose a threat to aviation safety if pilots—or even aircraft AIs—were to detect them in an irrational way (e.g., as alien objects). A novel theorization that integrates major perception, clinical, and cognition models is offered. Additionally, the authors discuss the implications for aviation safety and determine when a UAP sighting may pose a real danger on a commercial flight. •For the first time, UAPs have been introduced into Airbus® simulators to analyze pilots' reactions.•Our model forecasts over 80% of the variance in UAP sightings.•We propose criteria based on causal illusions to assess how pilots' UAP decisions risk aviation.