DeepSentry Platform: An Asymmetric AI System for Misinformation Detection

Abstract: Deployment of deep learning–based misinformation detection systems frequently suffersfrom severe resource contention when heterogeneous computational workloads share the same hardware. CPU-intensive preprocessing operations, such as video decoding and face extraction, interferewith GPU-intensive neural network inference, leading to GPU underutilization and Head-of-Lineblocking in monolithic architectures. This paper introduces DeepSentry, a microservice-based detection platform designed to mitigate this bottleneck through asymmetric architectural decoupling.In this context, asymmetry is defined as a non-uniform architectural decomposition in which CPUbound preprocessing services and GPU-bound inference services are intentionally isolated and scaledindependently, rather than replicated symmetrically as identical service instances. By separatingCPU-bound ingestion services from GPU-bound inference engines via message queuing and an inference server, the proposed platform enables independent scaling and asynchronous execution ofheterogeneous resources. A proof-of-concept implementation integrating ResNet34 for image deepfake detection with heatmap generation and BiLSTM for fake news classification demonstrates thatasymmetric microservice architectures significantly improve hardware utilization through dynamicbatching and asynchronous orchestration, effectively resolving the resource contention inherent inmonolithic deployments.Keywords: Resource contention, GPU starvation, Microservices, NVIDIA Triton, Deepfakedetection, Fake news detection.