Archive

This website uses cookies

Read our Privacy policy and Terms of use for more information.

Refusal Is a Setting. Detection Is the Control.

Jul 23, 2026

An open-weight model refused to help with an attack. A short reword got it to comply. The refusal was never the safeguard you thought it was.

Securing AI, Explained as a Building You Already Know How to Protect

Jul 14, 2026

A field guide to shadow AI, agentic gateways, and the master keys nobody meant to hand out.

Your AI Refused the Attack. Reworded, It Got Through.

Jul 8, 2026

We built a weak AI app on purpose, turned off every protection, and attacked it 816 times. Here is what got through, and why a reworded attack beats a direct one.

Your AI Guardrail Is On. The Attacker Already Got Past It.

Jun 24, 2026

A nine-attack lab on a real agentic LLM app: why enabled defenses still fall, mapped to OWASP and NIST CSF 2.0, and what to build instead.

Why Checking One Prompt at a Time Isn't Enough to Stop a Real Attack.

Jun 17, 2026

Most AI safety filters check each message on its own. The attacks that succeed spread a harmful request across many messages, or remove the safety directly from an open model's weights. Here is how they work, and what to do about it.

To poison your AI, an attacker no longer has to breach you. They just publish.

Jun 14, 2026

The open web your AI reads is an attack surface most teams are not watching. Here is the gap, shown with one live search, and how to close it this week.