The Global Navigation Satellite System (GNSS) poses a significant threat to autonomous drone (UAV) swarms. Although extensive research has concentrated on spoofing individual drones, the systemic vulnerabilities inherent in cooperative multi-agent configurations remain insufficiently examined. This paper offers a comprehensive, swarm-centric review of the GNSS spoofing threat, addressing a notable gap in the existing literature. We contend that despite advances in detection methodologies, a critical mitigation gap persists, namely the interval between attack detection and the implementation of effective real-time countermeasures. A salient feature of our study is the introduction of the inaugural hierarchical taxonomy, which categorises spoofing attacks based on the technical method, complexity, and exploitation of swarm behaviours. The paper analyses the literature concerning swarm-specific vulnerabilities, where an attack’s impact is exacerbated by the swarm’s control logic, resulting in cascading failures. Our comparative review of state-of-the-art solutions underscores the continuous trade-off between detection accuracy, computational load, and latency. Ultimately, we propose a research trajectory aimed at bridging the identified mitigation gap, emphasising the necessity of proactive AI-driven defences, robust Position, Navigation, and Timing (PNT) through multisensor integration, and the development of a theoretical framework for assessment to ensure the secure operation of future autonomous swarm systems.