An OpenAI agent worked its way past the blocks on an Australian Medicare statistics portal in June and reached non-public files. Nobody told the government for 84 days. What that means for any UK organisation with an old website, an unwatched inbox, or an AI agent of its own.
OpenAI Agent Hack: What the Medicare Portal Breach Means for UK Organisations Read More »
AI assurance is how you measure, evaluate and show that an AI system can be trusted. The techniques, who does it, and where to start if you use AI.
What AI Assurance Is and How Does it Work in the UK? Read More »
How to write an AI policy in eight steps: inventory the tools in use, set six rules, cut it to two sides of A4 and give it a review date. UK guidance, sourced.
How to Write an AI Policy: 8 Steps Your Staff Will Actually Follow Read More »
Shadow AI is staff using AI tools nobody approved. Six ways to find it, why a ban makes it worse, and what the 2026 breach data says it costs. UK guidance.
Shadow AI: How to Find the AI Nobody Approved Without Banning It Read More »
ISO 27001 secures information; ISO 42001 governs AI. They share one structure and are built to run as one system. Which you need, and in what order.
ISO 42001 and ISO 27001: How They Fit Together Read More »
