typo

robvanderveer · web-flow · commit 26aa3e7e7a13 · 2026-02-15T23:47:02.000+01:00
diff --git a/content/ai_exchange/content/docs/ai_security_overview.md b/content/ai_exchange/content/docs/ai_security_overview.md
@@ -9,35 +9,32 @@ weight: 1
 >Category: discussion  
 >Permalink: https://owaspai.org/goto/toc/
 
-* [AI Security Overview](docs/ai_security_overview/)  
-&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;- [About the AI Exchange](https://owaspai.org/goto/about/)  
-&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;- [Organize AI](https://owaspai.org/goto/organize/)  
-&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;- [How to use this document](https://owaspai.org/goto/document/)  
-&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;- [Essentials](https://owaspai.org/goto/essentials/)  
-&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;- [Threats](https://owaspai.org/goto/threatsoverview/)  
-&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;[Highlight: Threat matrix](https://owaspai.org/goto/aisecuritymatrix/)  
-&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;[Highlight: Agentic AI perspective](https://owaspai.org/goto/agenticaithreats/)  
-&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;- [Controls](https://owaspai.org/goto/controlsoverview/)  
-&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;[Highlight: Periodic table of threats and controls](https://owaspai.org/goto/periodictable/)  
-&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;- [Risk analysis](https://owaspai.org/goto/riskanalysis/)  
-&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;- [How about ...](https://owaspai.org/docs/ai_security_overview/#how-about-)  
-
-* [Deep dive into threats and controls:](https://owaspai.org/goto/navigator/)  
+* [AI Security Overview](/docs/ai_security_overview/)  
+&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;- [About the AI Exchange](/goto/about/)  
+&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;- [Organize AI](/goto/organize/)  
+&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;- [How to use this document](/goto/document/)  
+&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;- [Essentials](/goto/essentials/)  
+&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;- [Threats](/goto/threatsoverview/)  
+&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;[Highlight: Threat matrix](/goto/aisecuritymatrix/)  
+&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;[Highlight: Agentic AI perspective](/goto/agenticaithreats/)  
+&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;- [Controls](/goto/controlsoverview/)  
+&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;[Highlight: Periodic table of threats and controls](/goto/periodictable/)  
+&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;- [Risk analysis](/goto/riskanalysis/)  
+&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;- [How about ...](/docs/ai_security_overview/#how-about-)  
+
+* [Deep dive into threats and controls:](/goto/navigator/)  
 &nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;- [1. General controls](/docs/1_general_controls)  
-&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;[1.1 Governance controls](https://owaspai.org/goto/governancecontrols/)  
-&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;[1.2 Data limitation](https://owaspai.org/goto/datalimit/)  
-&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;[1.3 Limit unwanted behaviour](https://owaspai.org/goto/limitunwanted/)  
+&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;[1.1 Governance controls](/goto/governancecontrols/)  
+&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;[1.2 Data limitation](/goto/datalimit/)  
+&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;[1.3 Limit unwanted behaviour](/goto/limitunwanted/)  
 &nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;- [2. Input threats and controls](/docs/2_threats_through_use/)  
-&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;[Highlight: Prompt injection protection](https://owaspai.org/goto/promptinjectionsevenlayers/)  
+&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;[Highlight: Prompt injection protection](/goto/promptinjectionsevenlayers/)  
 &nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;- [3. Development-time threats and controls](/docs/3_development_time_threats/)  
 &nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;- [4. Runtime conventional security threats and controls](/docs/4_runtime_application_security_threats/)  
 
 * [AI security testing](/docs/5_testing/)
-
 * [AI privacy](/goto/aiprivacy/)  
-
 * [References](/docs/ai_security_references/)
-
 * [Index](/docs/ai_security_index/)
 
 ## About the AI Exchange
@@ -578,7 +575,7 @@ Discovering potential risks that could impact the organization requires the tech
  
   Question: Do you use RAG (Retrieval Augmented Generation) ?
   Yes: Then your retrieval repository plays a role in determining the model behaviour. This means:
-  - You need to protect against [leaking](/goto/leakaugmentation/) or [manipulation](/goto/manipulateaugmentation/) of your augmentation data (e.g., vector database), which includes preventing that it contains externally obtained poisoned data.
+  - You need to protect against [leaking](/goto/augmentationdataleak/) or [manipulation](/goto/augmentationdatamanipulation/) of your augmentation data (e.g., vector database), which includes preventing that it contains externally obtained poisoned data.
 
   Question: Who runs the model?
   - The supplier: select a trustworthy supplier through [supply chain management](/goto/supplychainmanage/), to make sure the deployed model cannot be manipulated ([runtime model poisoning](/goto/runtimemodelpoison/)) - just the way you would expect any supplier to protect their running application from manipulation.
