By name: magiq-post-quantum-agent-governance description: "Multi-agent AI governance system with provable post-quantum security (MAGIQ framework). Defines and enforces communication/access-control policies for agent-to-agent sessions using quantum-resistant cryptography with UC security proofs. Use when: post-quantum agent security, multi-agent governance, quantum-resistant multi-agent systems, agent accountability, SAGA framework comparison, UC framework multi-agent security, NIST PQC agent policies. Triggered by: MAGIQ, post-quantum agentic AI, quantum-resistant agent governance, multi-agent cryptographic security, agent communication policy enforcement, UC framework agent security, NIST post-quantum agent systems."
MAGIQ: Post-Quantum Multi-Agent AI Governance
Policy definition and enforcement framework for multi-agent AI systems using quantum-resistant cryptographic protocols with provable security guarantees.
Paper
arXiv: 2605.06933v1 — MAGIQ: A Post-Quantum Multi-Agentic AI Governance System with Provable Security by Sepideh Avizeh, Tushin Mallick, Alina Oprea, Cristina Nita-Rotaru, Reihaneh Safavi-Naini (May 2026).
Core Capabilities
- Policy Definition: Rich communication and access-control policy budgets for agent-to-agent sessions, including global one-to-many budgets.
- Post-Quantum Enforcement: Policies enforced using PQC primitives (NIST-recommended, replacing RSA/DH/ECC by 2035).
- Session-Based Enforcement: Agent-to-agent and one-to-many session policies.
- Accountability: Message attribution ensuring agents are accountable to users.
Security Model
- Formal correctness and security proofs using the Universal Composability (UC) framework
- Computation and communication overhead evaluated against SAGA baseline
- First post-quantum-secure solution for agentic AI systems
When to Use
- Designing secure multi-agent AI architectures
- Implementing post-quantum cryptography for agent communication
- Building accountability mechanisms for agentic systems
- Comparing governance frameworks (SAGA vs MAGIQ)