Resource Center

AI Security Research,
Guides & Tools

Deep research, real incident analyses, and practical guides for governing autonomous AI — from the RuntimeAI security team and independent analysts.

Analyst Research

Framework · Stanford + MVP Ventures

SACR Framework — Independent AI Security Analysis

Stanford GSB + MVP Ventures researchers independently described what RuntimeAI built. Three governance layers. 15 vendors analyzed. RuntimeAI: 10/10 — the only vendor covering all three layers.

Read Framework →

Analyst Reports

Third-Party Validation & Market Research

Independent analyst coverage and market research validating RuntimeAI's approach to AI agent governance.

View Reports →

Security Guides

AI Agent Security — Best Practices Guide

Comprehensive guide to governing autonomous AI agents: identity, policy, enforcement, monitoring, and incident response across enterprise deployments.

PQC Readiness Guide for AI Infrastructure

Post-quantum cryptography transition guide for AI systems. ML-KEM, ML-DSA, and SPHINCS+ implementations for agent identity, signing, and in-transit protection.

MCP Security — Securing Model Context Protocol

53% of public MCP servers have hardcoded secrets. This guide covers MCP threat model, SPIFFE mTLS for MCP, supply chain verification, and shadow server detection.

Incident Analysis & Research

Incident Analysis

Vercel AI SDK Incident — What Broke and Why

Deep-dive analysis of the Vercel AI SDK incident. What governance controls were absent, how RuntimeAI's enforcement layer would have caught it, and lessons for the industry.

Read Analysis →

Threat Research

LiteLLM Supply Chain Attack — MCP Threat Taxonomy

How attackers are targeting AI integration layers. Complete threat taxonomy for MCP supply chain attacks and mitigation strategies using the AI Integration Fabric.

Read Research →

AI Datacenters — The Governance Gap

As AI factories scale to thousands of agents on GPU infrastructure, the missing layer is governance. What needs to be built above the compute provisioning layer.

Read Research →

Physical AI Governance — Robots Need Kill Switches Too

Applying AI agent governance to physical systems: robots, drones, and autonomous vehicles. Hardware attestation, geofencing, and emergency stop protocols.

Read Research →

Quantum Day 2026 — Post-Quantum Cryptography for AI

NIST PQC standards are final. What does ML-KEM and ML-DSA mean for AI agent identity, signing, and in-transit security? The complete migration guide.

Read Research →

Weekly AI security digest, incident analyses, research papers, product updates, and governance thought leadership from the RuntimeAI team.