Quantum-Safe Networking for Critical AI/ML Infrastructure

Abstract

The integration of artificial intelligence (AI) and machine learning (ML) into critical infrastructure has raised urgent concerns about data and model security, particularly in light of emerging quantum computing capabilities. Quantum algorithms threaten to render classical cryptographic methods obsolete, exposing AI/ML systems to potential breaches in confidentiality, integrity, and availability. This paper investigates the implications of quantum computing for securing AI/ML data both in transit and at rest and explores the development of quantum-safe networking protocols and cryptographic techniques.

I examine post-quantum cryptographic (PQC) solutions including lattice-based, code-based, and hash-based algorithms, alongside the role of quantum key distribution (QKD) and AI-enhanced security orchestration. The study further addresses secure edge intelligence, federated AI systems, and emerging standards for 6G and beyond.

My findings highlight both the necessity and complexity of transitioning to quantum-resilient infrastructure. Key challenges include computational overhead, legacy interoperability, and ethical concerns around AI-powered surveillance in quantum-secured environments. The paper concludes by emphasizing the need for proactive policy, investment in quantum-safe R&D, and cross-sector collaboration to safeguard AI/ML infrastructure in the post-quantum era.