Personal publication

Clear writing on messy security problems.

Notes for engineers who need decisions, not fear: application security, AI systems, threat modeling, and what happens when incentives meet real attackers.

Read the blog About the author

Featured

Deeper dives worth starting with.

Full archive →

Mar 25, 20265 min readFeatured

Tokenization Attacks: The AI Security Risk Almost Nobody Models

Why retrieved text is not just ‘untrusted input’ but an attacker-controlled instruction channel when your model can call tools — and how to think about controls without magical safety guarantees.

AI Security
LLM
Offensive Security
AppSec

Topics

Jump in by theme — same tags as on each post.

AI Security
LLM
Offensive Security
AppSec

Browse all tags →