The progression of fifth-generation (5G) networks provides multiple advantages, such as faster speed, reduced latency and increased capacity. Towards these advancements, it is clear that 5G Core (5GC) represents the heart of a 5G network, providing a variety of new network services such as Ultra-reliable low-latency communication (URLLC) and Massive machine-type communication (mMTC). However, despite the various benefits, 5GC is prone to several cyberthreats that can result in catastrophic effects. In this paper, an Intrusion Detection System (IDS) for 5GC is introduced. The proposed IDS called 5GCIDS adopts Artificial Intelligence (AI) methods in order to detect potential cyberattacks against Packet Forwarding Control Protocol (PFCP), which is utilised for the N4 interface between Session Management Function (SMF) andUser Plane Function (UPF). For the detection process, both Transmission Control Protocol (TCP)/Internet Protocol (IP) flow statistics and application-layer PFCP flow statistics are used. In the second case, we provide a bidirectional flow statistics generator called PFCPFlowMeter. Finally, the detection outcomes are explained as local and global explanations with the TreeSHAP method. The evaluation results demonstrate the efficiency of the proposed IDS.