<?xml version="1.0" encoding="UTF-8"?><rss version="2.0"><channel><title>Encipher Trust — Blog</title><description>Analysis and practical guidance on cybersecurity, AI security, cryptography and financial technology.</description><link>https://enciphertrust.com/</link><language>en</language><item><title>Harvest now, decrypt later: why post-quantum planning starts with an inventory</title><link>https://enciphertrust.com/resources/blog/harvest-now-decrypt-later-pqc-inventory/</link><guid isPermaLink="true">https://enciphertrust.com/resources/blog/harvest-now-decrypt-later-pqc-inventory/</guid><description>NIST has standardised the first post-quantum algorithms. The hardest part of migration is not the algorithms — it is knowing where your cryptography is.</description><pubDate>Tue, 18 Aug 2026 00:00:00 GMT</pubDate><category>Post-Quantum Cryptography</category><category>PQC</category><category>Cryptography</category><category>Risk Management</category><author>contact@enciphertrust.com (Encipher Trust Research)</author></item><item><title>Explainable credit decisions: designing reason codes that customers and regulators can trust</title><link>https://enciphertrust.com/resources/blog/explainable-credit-decisions-reason-codes/</link><guid isPermaLink="true">https://enciphertrust.com/resources/blog/explainable-credit-decisions-reason-codes/</guid><description>A credit model is only as deployable as its explanations. How to design reason codes that are accurate, consistent and useful.</description><pubDate>Wed, 22 Jul 2026 00:00:00 GMT</pubDate><category>AI Underwriting</category><category>Explainability</category><category>Credit Risk</category><category>Machine Learning</category><author>contact@enciphertrust.com (Encipher Trust Research)</author></item><item><title>Indirect prompt injection: the security risk hiding in your AI assistant’s documents</title><link>https://enciphertrust.com/resources/blog/indirect-prompt-injection-ai-assistants/</link><guid isPermaLink="true">https://enciphertrust.com/resources/blog/indirect-prompt-injection-ai-assistants/</guid><description>When an LLM reads untrusted content, that content can attempt to give it instructions. Why retrieval and tools change the threat model.</description><pubDate>Tue, 30 Jun 2026 00:00:00 GMT</pubDate><category>AI Security</category><category>LLM Security</category><category>AI Security</category><category>Application Security</category><author>contact@enciphertrust.com (Encipher Trust Research)</author></item></channel></rss>