Defensive AI Security & Governance
Building the discipline of autonomous defense — detecting drift, governing intent, and containing AI threats at machine speed, with the human kept where judgment matters.
Research-grade analysis of AI security — grounded in public, authoritative sources. Watch the narrated walkthrough of every topic.
Building the discipline of autonomous defense — detecting drift, governing intent, and containing AI threats at machine speed, with the human kept where judgment matters.
Attack & defend AI systems — threat labs, primitives and chains, every offensive step paired with its defense.
Rare attacks that break unnamed assumptions — like systems disagreeing on what the same bytes mean — each paired with its defense.
The post-quantum transition beyond the lattice — the code-based and multivariate alternatives, what they cost, and where the standards stand.
Computer-use agents that see and act — when the screen becomes untrusted input, and how to contain the damage.
What a model can really do — sandbagging, eval awareness, reward hacking and scheming, measured honestly.
The algorithms behind a model's answer — attention, the KV cache, speculative decoding, quantization, and vector recall — derived properly, to where each one breaks.
Reading model internals as a security control — probes, features and their limits.
How one engineer and an AI assistant build end-to-end — argued from evidence.
What a clean static-analysis report can and cannot prove about AI-agent software.