Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

When AI Fails, What Works? A Data-Driven Taxonomy of Real-World AI Risk Mitigation Strategies

Evgenija Popchanovska, Ana Gjorgjevikj, Maryan Rizinski, Lubomir Chitkushev, Irena Vodenska, Dimitar Trajanov

TL;DR

By structuring incident responses, the paper strengthens diagnosis-to-prescription guidance and advances continuous, taxonomy-aligned post-deployment monitoring to prevent cascading incidents and downstream impact and enhances the taxonomy's applicability to emerging systemic failure modes.

Abstract

Large language models (LLMs) are increasingly embedded in high-stakes workflows, where failures propagate beyond isolated model errors into systemic breakdowns that can lead to legal exposure, reputational damage, and material financial losses. Building on this shift from model-centric risks to end-to-end system vulnerabilities, we analyze real-world AI incident reporting and mitigation actions to derive an empirically grounded taxonomy that links failure dynamics to actionable interventions. Using a unified corpus of 9,705 media-reported AI incident articles, we extract explicit mitigation actions from 6,893 texts via structured prompting and then systematically classify responses to extend MIT's AI Risk Mitigation Taxonomy. Our taxonomy introduces four new mitigation categories, including 1) Corrective and Restrictive Actions, 2) Legal/Regulatory and Enforcement Actions, 3) Financial, Economic, and Market Controls, and 4) Avoidance and Denial, capturing response patterns that are becoming increasingly prevalent as AI deployment and regulation evolve. Quantitatively, we label the mitigation dataset with 32 distinct labels, producing 23,994 label assignments; 9,629 of these reflect previously unseen mitigation patterns, yielding a 67% increase of the original subcategory coverage and substantially enhancing the taxonomy's applicability to emerging systemic failure modes. By structuring incident responses, the paper strengthens "diagnosis-to-prescription" guidance and advances continuous, taxonomy-aligned post-deployment monitoring to prevent cascading incidents and downstream impact.

When AI Fails, What Works? A Data-Driven Taxonomy of Real-World AI Risk Mitigation Strategies

TL;DR

By structuring incident responses, the paper strengthens diagnosis-to-prescription guidance and advances continuous, taxonomy-aligned post-deployment monitoring to prevent cascading incidents and downstream impact and enhances the taxonomy's applicability to emerging systemic failure modes.

Abstract

Large language models (LLMs) are increasingly embedded in high-stakes workflows, where failures propagate beyond isolated model errors into systemic breakdowns that can lead to legal exposure, reputational damage, and material financial losses. Building on this shift from model-centric risks to end-to-end system vulnerabilities, we analyze real-world AI incident reporting and mitigation actions to derive an empirically grounded taxonomy that links failure dynamics to actionable interventions. Using a unified corpus of 9,705 media-reported AI incident articles, we extract explicit mitigation actions from 6,893 texts via structured prompting and then systematically classify responses to extend MIT's AI Risk Mitigation Taxonomy. Our taxonomy introduces four new mitigation categories, including 1) Corrective and Restrictive Actions, 2) Legal/Regulatory and Enforcement Actions, 3) Financial, Economic, and Market Controls, and 4) Avoidance and Denial, capturing response patterns that are becoming increasingly prevalent as AI deployment and regulation evolve. Quantitatively, we label the mitigation dataset with 32 distinct labels, producing 23,994 label assignments; 9,629 of these reflect previously unseen mitigation patterns, yielding a 67% increase of the original subcategory coverage and substantially enhancing the taxonomy's applicability to emerging systemic failure modes. By structuring incident responses, the paper strengthens "diagnosis-to-prescription" guidance and advances continuous, taxonomy-aligned post-deployment monitoring to prevent cascading incidents and downstream impact.
Paper Structure (10 sections, 2 figures, 1 table)

This paper contains 10 sections, 2 figures, 1 table.

Table of Contents

  1. Introduction
  2. Literature Review
  3. Data and Methodology
  4. Data Collection
  5. Methodology
  6. Taxonomy
  7. Incident Categorization
  8. Industry and financial impact of AI incidents
  9. Conclusion and Future Work
  10. Extended AI Risk Mitigation Taxonomy (only new categories and subcategories shown)

Figures (2)

  • Figure 1: Distribution of subcategory labels across the dataset mitigation actions, showing 9,629 labels for newly identified subcategories and 14,365 labels matching existing subcategories. The new and existing categories are marked orange and dark blue, respectively.
  • Figure 2: Frequency distribution across existing and newly introduced subcategory labels. The new labels reveal previously unseen mitigation patterns, representing a 67% expansion of the original subcategories.