SETONacademy · Cybersecurity
Short, practical briefings on the threats parish staff actually encounter — phishing, endpoint hygiene, network risk, AI prompt injection, incident response. The Cybersecurity Minute series is live now. More follows.
Deeper detail for the Cybersecurity Minute series above. These are the three patterns that account for most prompt-injection incidents in office settings.
A PDF, a Word doc, or a webpage contains text — sometimes invisible white-on-white — telling the AI to ignore your instructions and do something else instead. Always treat AI output from external documents as a draft, not a finished product.
An email arrives with instructions buried in the signature or quoted history. When you ask Copilot to summarize or reply, it reads those instructions too. Be cautious with AI assistance on threads that include unknown senders.
An attacker's instructions tell the AI to render an image hosted on their server, with sensitive information encoded in the URL. The AI "renders" the image, the server logs the URL, and the data has left the building.
The AI Prompt Injection sub-series is the first. The next briefings record across 2026.
What the modern phishing email actually looks like, why it gets through filters, and the two checks that catch most of it before damage is done.
Laptops, phones, parish-office desktops. Patching, password managers, MFA, and what to do when a device walks out the door.
Incident response for parishes. Who to call, what to record, what not to do in the first hour, and how to talk about it after.
A one-page talking outline you can adapt to your next staff meeting. Plain language. The patterns to flag, the habits to build.
Cybersecurity is one topic. The full curriculum covers AI, networking, physical security, data protection, and compliance.
Back to the hub