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

Building a Production AI Stack — Vector DB, LLM, Auth, Observability

A real production AI application has 6-8 components: LLM (own or API), embedding model, vector DB, prompt cache, auth, rate limit, content…

Apr 29, 2026 · 9 min read
AI Security

Defending AI Endpoints — Rate Limit, Content Filters, NeMo Guardrails, Llama Guard

Once your AI endpoint is public, attackers will probe it within hours — for free LLM access, prompt injection, content-policy violations, and…

Apr 29, 2026 · 8 min read
AI Security

AI Security 101 — Why ML Systems Break Differently

Traditional software is deterministic. ML systems are probabilistic, learn from data, and respond to natural language. That changes the entire threat model…

Apr 29, 2026 · 10 min read
AI Security

Multi-Modal Attacks — Image Prompt Injection and Audio Adversarials

GPT-4V, Claude 3.5 Sonnet, and Gemini accept images. Whisper, ElevenLabs, and others accept audio. Each modality is an injection surface. This module…

Apr 29, 2026 · 8 min read
AI Security

Prompt Injection — Direct, Indirect, and Why It Will Not Be Patched

Prompt injection is to LLMs what SQL injection was to web apps in 2002 — except this time there is no equivalent…

Apr 29, 2026 · 10 min read
AI Security

Browser-Use Agents — Risks When LLMs Browse the Web

Anthropic computer-use Claude, OpenAI Operator, and frameworks like browser-use let agents control real browsers. They click, type, fill forms, log in. Every…

Apr 29, 2026 · 9 min read
AI Security

Data Poisoning and AI Supply Chain — Attacks Before Deployment

Most AI defenders worry about runtime attacks. Sophisticated attackers go upstream — poisoning training data, hijacking model registries, planting backdoors in fine-tuned…

Apr 29, 2026 · 9 min read
AI Security

AI Supply Chain — Hugging Face Hijacks, Pickle Attacks, Model Card Poisoning

You download a model from Hugging Face. The model file format (Pickle) supports arbitrary code execution on load. The model card lies…

Apr 29, 2026 · 8 min read
1 7 8 9