AI Security in Practice

Deep technical implementation guides for securing AI systems in production. The practitioner layer between vendor marketing and unreadable PDFs.

Read the Manifesto GitHub

Recently Added

Emerging Threats4 Aug 2026Quantum Threats to AI Governance: Why the Audit Logs You Build Today Can Be Forged TomorrowAI governance frameworks are being built on cryptographic foundations the cryptographic community has already declared will eventually fail. A technical analysis of the quantum threat to AI audit trails, agent trust chains, and compliance infrastructure.

Explore by Pillar

FoundationsStart here. LLM internals, OWASP Top 10, tool selection, and your first security lab.

Attack and Red TeamPrompt injection, jailbreaking, red teaming tools, adversarial ML, and CTFs.

Defend and HardenGuardrails, secure RAG, output validation, MCP security, and agent hardening.

Architecture and PlatformCloud AI security, MLOps, API gateway patterns, and Zero Trust for AI.

Governance, Risk and ComplianceFrameworks, compliance, risk registers, incident response, and UK Government guidance.

Emerging Threats and ResearchAI worms, sleeper agents, supply chain attacks, and multi-agent system security.