---
title: "Comment by Kartik N V J K on \"Securing AI Memory Against Poisoning in .NET\" | Hashnode"
description: "The split between a write-time gate and a retrieval-time gate is the part most designs skip, since people tend to sanitize on the way in and then trust whatever"
url: "https://hashnode.com/posts/securing-ai-memory-against-poisoning-in-net-applications/6a553e4a0b1d5f687a88eed0/comment/6a5664720eed3878e7a03ba9"
---

[Discussion](https://hashnode.com/posts/securing-ai-memory-against-poisoning-in-net-applications/6a553e4a0b1d5f687a88eed0)

Comment [Securing AI Memory Against Poisoning in .NET](https://hashnode.com/posts/securing-ai-memory-against-poisoning-in-net-applications/6a553e4a0b1d5f687a88eed0)

[Kartik N V J K](https://hashnode.com/@kartik-nvj)

[Jul 14](https://hashnode.com/posts/securing-ai-memory-against-poisoning-in-net-applications/6a553e4a0b1d5f687a88eed0/comment/6a5664720eed3878e7a03ba9)

The split between a write-time gate and a retrieval-time gate is the part most designs skip, since people tend to sanitize on the way in and then trust whatever comes back out. I like that provenance and expiry travel with each entry, because a poisoned fact that survives across sessions is the failure mode nobody sees until a tool acts on it. How are you deciding trust downgrades over time, is it fixed expiry or does retrieval frequency factor in?
