A personal perspective on security work: what we think, what we have learned and where we disagree with the consensus. The newsroom reports; the blog argues.
A practical look at prompt injection attacks on AI systems — direct and indirect injection, RAG poisoning, AI agent risks — and the enterprise security controls that actually reduce them.
What this blog is for, how it differs from the newsroom and Insights, and what to expect here.
News reports what happened, with sources. Insights are editorial explainers in the newsroom voice. The blog is opinion: it says what we think and why, and it is labelled that way.