From ea34074456b553f5ae5410ec51706d1d174a5790 Mon Sep 17 00:00:00 2001 From: Matteo Meucci Date: Thu, 20 Nov 2025 17:33:25 +0100 Subject: [PATCH] Update 2.1.2_Identify_RAI_threats.md --- .../content/2.1.2_Identify_RAI_threats.md | 24 +++++++++---------- 1 file changed, 12 insertions(+), 12 deletions(-) diff --git a/Document/content/2.1.2_Identify_RAI_threats.md b/Document/content/2.1.2_Identify_RAI_threats.md index 66d9443..ea8c415 100644 --- a/Document/content/2.1.2_Identify_RAI_threats.md +++ b/Document/content/2.1.2_Identify_RAI_threats.md @@ -1,10 +1,10 @@ -# **2.1.2 Identify AI System RAI and Trustworthy threats** +# **2.1.2 Identify AI System RAI threats** ## **Expanding Threat Modeling Beyond Security** Modern AI systems are not only susceptible to technical security threats like prompt injection or data poisoning, but also to ethical, social, and governance risks that affect trustworthiness, fairness, and compliance. Threat Modeling performed in the previous paragraph is not enough, we need to expand the approach. -Before we review it, let's briefly revisit key Trustworthy AI concepts defined by NIST and European Commission guidelines: +Before we review it, let's briefly revisit key Responsible AI concepts defined by NIST and European Commission guidelines: * Explainability & Transparency: AI outputs must be understandable to users. * Fairness & Non-discrimination: AI must avoid bias and discriminatory outcomes. @@ -21,7 +21,7 @@ Based on these concepts, we carefully revise your previous identification of thr *Fig. 3 Responsible AI Threat modeling* -### **Responsible AI / Trustworthy AI Additions Only** +### **Responsible AI Additions Only** #### **🟦 AI Application (5 additions):** @@ -47,30 +47,30 @@ Based on these concepts, we carefully revise your previous identification of thr 11. Toxic or Harmful Training Data (T01-T&HD) 12. Data Privacy Violations (T01-DPV) -The following tables provide a structured overview, facilitating clear visibility for identifying, managing, and mitigating critical Responsible AI and Trustworthy AI (RT) threats across your entire AI system architecture. +The following tables provide a structured overview, facilitating clear visibility for identifying, managing, and mitigating critical Responsible AI (RAI) threats across your entire AI system architecture. -## **🟦 AI Application RT Threats** +## **🟦 AI Application RAI Threats** -| \# | Component | Responsible & Trustworthy AI Threats | +| \# | Component | Responsible AI Threats | | ----- | ----- | ----- | | 1 | Application & User Interaction | Hallucinations, Toxic Content Generation, Automation Bias, Lack of Explainability | | 2 | Agent/Plugin | Lack of Accountability in Agent/Plugin Actions | -## **🟪 AI Model RT Threats** +## **🟪 AI Model RAI Threats** -| \# | Component | Responsible & Trustworthy AI Threats | +| \# | Component | Responsible AI Threats | | ----- | ----- | ----- | | 3 | Model (Inference & Training) | Bias and Discrimination, Overfitting and Generalization, Misalignment with Human Intent | -## **🟩 AI Infrastructure RT Threats** +## **🟩 AI Infrastructure RAI Threats** -| \# | Component | Responsible & Trustworthy AI Threats | +| \# | Component | Responsible AI Threats | | ----- | ----- | ----- | | 4 | Infrastructure (Storage, Serving, Frameworks, Training & Evaluation) | Lack of Traceability & Auditability of Infrastructure Processes | -## **🟨 AI Data RT Threats** +## **🟨 AI Data RAI Threats** -| \# | Component | Responsible & Trustworthy AI Threats | +| \# | Component | Responsible AI Threats | | ----- | ----- | ----- | | 5 | Data (Sources, Processing, Storage) | Non-representative Training Data, Toxic or Harmful Training Data, Data Privacy Violations |