AI Security · 88 articles

AI Security

AI / LLM security — prompt injection, data poisoning, agent threat models, building trustworthy AI, optimisation, and the architecture of trending AI products.

AI Security

PyRIT: Microsoft’s Python Risk Identification Tool for Generative AI

PyRIT gives red-teamers a programmable framework — orchestrators, prompt converters, and scorers — to automate generative-AI attacks at scale.

May 25, 2026 · 2 min read
AI Security

Mindgard and the Rise of AI-Native Offensive Security Platforms

A new category arrived in 2026: AI-native offensive security platforms that test models, agents, and multimodal apps continuously, inside the SDLC.

May 25, 2026 · 2 min read
AI Security

Garak: NVIDIA’s LLM Vulnerability Scanner — A Practitioner’s Guide

Garak probes LLMs for prompt injection, jailbreaks, toxicity, and data leakage with hundreds of built-in attack probes. A hands-on primer.

May 25, 2026 · 3 min read
AI Security

Promptfoo for Red Teamers: Automated GenAI Attack Testing in Your Pipeline

Promptfoo turns LLM red-teaming into repeatable, CI-friendly test suites — prompt injection, data leaks, and jailbreaks on every build.

May 25, 2026 · 2 min read
AI Security

Securing the AI Supply Chain in 2026: Model Provenance, Poisoning, and MCP Tool Risk

Your AI is only as trustworthy as the weights, data, and tools behind it. The 2026 guide to AI supply-chain risk.

May 25, 2026 · 2 min read
AI Security

AI Governance in 2026: Early Model Access for Regulators, the EU AI Act, and DPDP × AI

Governments are pushing pre-release testing and early model access. For Indian teams, AI governance now sits at the intersection of DPDP and…

May 25, 2026 · 1 min read
AI Security

Shadow AI Is Your Biggest Data-Leak Risk in 2026 — and DPDP Makes It Expensive

Unsanctioned AI tools are now the #1 data-leak entry point. Under the DPDP Act, that is not just a security problem —…

May 25, 2026 · 2 min read
AI Security

AI Red-Teaming Goes Autonomous: Agents That Hack Other AIs

Autonomous agents now solve most black-box red-team challenges faster than humans. What that means for offensive security in 2026.

May 25, 2026 · 2 min read
AI Security

Second-Order Prompt Injection: How Attackers Hijack Multi-Agent Systems

A new attack class: feed a low-privilege agent a malformed request so it convinces a higher-privilege agent to act. Here is how…

May 25, 2026 · 1 min read
AI Security

Prompt Injection Is Officially the #1 AI Risk — OWASP’s 2026 Agentic Top 10 Explained

OWASP keeps prompt injection at LLM01 and adds an Agentic Top 10 for 2026. A plain-English tour of the risks that matter.

May 25, 2026 · 2 min read