Air Traffic Controllers (ATCOs) play a vital role in maintaining the safety and efficient functioning of air traffic control (ATC) systems. To execute their responsibilities effectively and uphold the safety and efficiency of air traffic management, ATCOs must sustain a high degree of situational awareness (SA). A decline in SA can result in operational errors, posing significant risks to aviation safety. Therefore, the purpose of this paper is to propose a real-time Transformer-enabled SA recognition method based on eye-tracking technology, effectively learning temporal information for ATCOs’ accurate SA monitoring. To validate the proposed method called ConFormer, an in-lab simulation experiment was conducted with 26 participants involved. The eye movements of participants during each trial were recorded using the eye tracker. Extensive evaluation was performed to validate the effectiveness and reliability of the proposed Transformer-enabled SA estimator. The results demonstrate that the proposed method achieved superior performance, with a recognition accuracy of 98.0%, precision of 97.9%, F1 score of 97.7%, and recall of 97.5%, significantly outperforming baseline models such as LSTM (Long Short-Term Memory). The proposed SA recognition method can significantly enhance human-AI collaboration and aviation safety. This capability allows adaptive automation systems to dynamically adjust support during periods of high cognitive workload, for example, prioritizing alerts when SA declines, while maintaining ATCOs’ supervisory roles.