News & insights
Practical writing on AI security, cyber risk and the systems we are building.
For anyone who buys, commissions or simply wonders about penetration testing. How the job changed between 2024 and 2026, why "anyone can be a pentester now" is only half true, and what we expect in 2027.
Read articleFor anyone whose company website runs on PHP, WordPress included. What AI found this week, why it reaches so far, and what it would have looked like if attackers had found it first and told nobody.
Read articleMalicious MCP servers already exist, and they pass your AI instructions you don't see. What they look like, why companies switch MCP off, and what to do instead.
Read articleFor anyone who uses ChatGPT, Claude, Gemini or Copilot at work and has wondered where the conversation goes once the window is closed. Real cases from 2025 and 2026 show where private data typed into AI chats ends up, and what not to paste.
Read articleFor companies with a chatbot that talks to customers. People have talked bots into agreeing to a $1 car and giving away a shop's stock, bots have invented refund rules on their own, and so far tribunals and courts have held the company responsible. Here's how the tricks work, why bots fall for them, and what stops them.
Read articleFor anyone who has downloaded a free AI model, or is about to. Researchers have shown this year how cheaply a model can be trained to leak data, and the usual checks won't catch it. Here is what's known so far, and what to check before you connect one to your files, your email or your work.
Read article
