Welcome to The Attack Surface

Threat intelligence, offensive security, and AI security: why this blog exists and what to expect.

Every system worth attacking has an attack surface: the sum of every door, window, and crack an adversary can probe. This blog is about mapping those surfaces: the ones defenders overlook, the ones attackers are actively working, and the entirely new ones that AI systems are opening up.

What you'll find here

Threat intelligence. Analysis of active campaigns, adversary tradecraft, and the infrastructure behind them, with an emphasis on what's actionable rather than what's merely interesting.

Offensive security. Techniques, tooling, and lessons from the attacker's side of the fence. Understanding how systems break is the only reliable way to understand how to defend them.

AI security. Large language models and agentic systems are becoming part of the attack surface: as targets, as tools, and as a new class of insider. Prompt injection, model supply chains, agent privilege escalation: this space is moving fast, and much of the conventional wisdom is already wrong.

Why "The Attack Surface"

Because that's where everything interesting happens. Defense-in-depth diagrams and compliance frameworks describe the world as we wish it were; the attack surface is the world as it actually is. That's the vantage point this blog writes from.

More soon. If you want posts in your inbox, subscribe below.

A futuristic robot head with a glowing red triangle symbol on a dark background

JADEPUFFER: The First Ransomware Attack an AI Ran by Itself

Sysdig documented a ransomware operation with no human at the keyboard. An LLM broke into an exposed Langflow server, moved laterally, and encrypted a production database start to finish. The tactics were old. The operator was not a person.

A translucent x-ray style rendering of a human brain on a black background

The Intent Layer: What J-Space Means for Agentic AI Security, and What It Doesn't

Anthropic's J-space research promises something like intent telemetry for AI agents. The opportunity is real, but it is not the logs we never had: it is a noisy, adversarially fragile sensor that needs an enforcement point you should have built anyway.

A black and yellow hazard sign bearing a skull symbol

The Lethal Trifecta, Explained With Real Incidents

Private data, untrusted content, and an exfiltration channel. When an agent has all three, someone else decides what it does with your secrets. Here is the pattern, drawn from three real breaches.

A toy robot sitting at a laptop keyboard

Your AI Agent Is an Insider Threat

AI agents have credentials, autonomy, and total gullibility. Stop securing them like software and start securing them like staff.