AI Security & Models — Issue #42 (Aug 17, 2026)
OpenAI's gated cyber model, GLM-5.3's CyberGym lead, real agent breakouts, EU watermarking rules, and Grok Bot's "coworker" beta—plus what to watch next.
Read more
AI audit readiness: a practical checklist for CISOs
AI is already in your estate. The only question is whether you discover it in an AI audit or an incident report.
Self Hosted AI vs ChatGPT: Why Governance Teams Choose Local LLMs
This article breaks down five reasons governance teams are pivoting to self hosted models over ChatGPT style APIs, and two alternatives most overlook: SaaS with BYOK AI backends, and regional providers that satisfy residency requirements.
AI Sovereignty Starts at the Inference Layer not the Border
Swiss banks understood the risks of foreign infrastructure. AI now shows why data sovereignty must cover inference, not just storage.
ISO 42001 vs EU AI Act for CISOs
A practical guide to where ISO 42001 helps, where the EU AI Act bites, and how to use both without building two separate programs.