A Zero-Trust AI-Blockchain Architecture for Quantum-Secure Metaverse Platforms

Abstract

The growth of the Metaverse brings new security problems that traditional perimeter-based defenses can’t manage. This research proposes and tests an integrated Zero-Trust Architecture aimed to solve these weaknesses by merging artificial intelligence (AI)-driven behavioral threat detection, blockchain-based decentralized identification, and post-quantum cryptography. For anomaly detection, the architecture uses a federated ResNet-50 model; for data management that meets regulatory standards, it uses a Hyperledger Fabric-based identification system with Zero-Knowledge Succinct Non-Interactive Argument of Knowledge; and for key exchange that is immune to quantum attacks, it uses the CRYSTALS-Kyber algorithm. Penetration testing, a Delphi study with 20 experts, and user surveys all show that the architecture greatly improves security metrics. This system has a False Acceptance Rate (FAR) of 5.2%, which is 42.7% lower than the 9.1% FAR baseline of rule-based systems, 99.1% protection against Sybil attacks, and strong quantum resilience with a 1.2× latency penalty compared to AES-256. The approach also partially complies with the General Data Protection Regulation by using cryptographic erasure proofs. But these security improvements come at a cost: AI inference now uses 3.1 times more graphics processing unit resources. The results show that the suggested architecture creates a scalable, empirically validated basis for protecting decentralized virtual environments, striking a good balance between security, compliance, and performance trade-offs.