https://piishield.ai/blog Keeping personal data out of AI, in practice. https://piishield.ai/blog/connect-ats-crm-to-claude-without-exposing-personal-datahttps://piishield.ai/blog/connect-ats-crm-to-claude-without-exposing-personal-dataTue, 14 Jul 2026 09:00:00 GMTAdding your ATS to Claude as a connector does not protect it — Claude talks to it directly and every name it receives is real. Here is what to put in between, and why one address for several systems matters. https://piishield.ai/blog/stop-staff-pasting-customer-data-into-chatgpthttps://piishield.ai/blog/stop-staff-pasting-customer-data-into-chatgptTue, 30 Jun 2026 09:00:00 GMTPolicy does not survive contact with a deadline. What works is masking the text in the chat box before it is sent — and restoring the real values in the reply, so nobody has a reason to route around it. https://piishield.ai/blog/redact-pii-before-the-openai-and-anthropic-apihttps://piishield.ai/blog/redact-pii-before-the-openai-and-anthropic-apiThu, 18 Jun 2026 09:00:00 GMTA gateway is a base-URL change: redaction happens on every request whether or not anyone remembered, which is the only version that holds when the codebase grows. https://piishield.ai/blog/redact-a-word-excel-or-pdf-before-uploading-to-aihttps://piishield.ai/blog/redact-a-word-excel-or-pdf-before-uploading-to-aiFri, 05 Jun 2026 09:00:00 GMTDeleting text from a PDF usually does not delete it. What format-preserving redaction has to get right, and why scanned pages need OCR before anything can be masked at all. https://piishield.ai/blog/find-what-already-leaked-into-ai-chatshttps://piishield.ai/blog/find-what-already-leaked-into-ai-chatsFri, 22 May 2026 09:00:00 GMTPrevention says nothing about the last eighteen months. Auditing existing chat history is the uncomfortable half — including shared links, which are public web pages.