---
title: "Discussion on \"Why blocking prompt injection is wrong — and what to do instead\" | Hashnode"
description: "Discussion on \"Why blocking prompt injection is wrong — and what to do instead\". Instead of blocking prompt injection attacks, MIRAGE deceives attackers with a fake AI persona — burning their tokens while logging everything."
url: "https://hashnode.com/posts/why-blocking-prompt-injection-is-wrong-and-what-to-do-instead/6a1037111f237623eae2963a"
---

[Feed](https://hashnode.com/feed) /

Discussion

# [Why blocking prompt injection is wrong — and what to do instead](https://brightgirl.hashnode.dev/why-blocking-prompt-injection-is-wrong-and-what-to-do-instead?utm_source=hashnode&utm_medium=feed)

Instead of blocking prompt injection attacks, MIRAGE deceives attackers with a fake AI persona — burning their tokens while logging everything.

May 22 · 3 min read · brightgirl.hashnode.dev  
Read the article

-   [#llm-security](https://hashnode.com/tag/llm-security)
-   [#ai](https://hashnode.com/tag/ai)
-   [#security](https://hashnode.com/tag/security)
-   [#ai-security](https://hashnode.com/tag/ai-security)
-   [#opensource](https://hashnode.com/tag/opensource)

## Written by

[Victoria Fonareva](https://hashnode.com/@brightgirl)

@brightgirl · Warsaw, Poland

  
[View profile](https://hashnode.com/@brightgirl)

Software Engineer at Yandex Fintech | Go Developer

Building payment infrastructure at Yandex. Focused on backend scalability, Go, and practical LLM applications. I value high-density information and efficient execution.

## Responses

Markdown supported · ⌘↵ to post

No responses yet.
