Saved in:
| Main Authors: | , , , , , , |
|---|---|
| Format: | Preprint |
| Published: |
2026
|
| Subjects: | |
| Online Access: | https://arxiv.org/abs/2604.21412 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866914536156561408 |
|---|---|
| author | Mengesha, Isaak Owen, Branwen Collins, Charlie Wong, Tina Mylius, Simon Slattery, Peter McGregor, Sean |
| author_facet | Mengesha, Isaak Owen, Branwen Collins, Charlie Wong, Tina Mylius, Simon Slattery, Peter McGregor, Sean |
| contents | Incident monitoring can drive safety improvements in high-reliability industries and population-scale technologies, but remains underdeveloped in AI governance. Public databases catalog thousands of AI incidents, but simple incident counts conflate media reporting propensity, system deployment ("exposure"), and harm frequency per unit exposure. We propose a methodological framework that accounts for these factors and calibrates confidence to available evidence in analyzing how AI incidents change over time. The framework comprises three components: a structured monitoring question that defines the scope of the analysis; a tiered estimation process that separately derives harm and exposure trends, including through LLM-assisted filtering of public incident databases; and a classification scheme that maps the resulting trend estimates onto actionable governance categories (Escalating, Mitigating, Concentrating, Receding or Unclassifiable). Through case studies, we examine the framework's clarifying power and limitations, demonstrate governance insight despite real-world data constraints, and provide a proof of concept for AI incident monitoring as a practical governance tool. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2604_21412 |
| institution | arXiv |
| publishDate | 2026 |
| record_format | arxiv |
| spellingShingle | A pragmatic classification framework for AI incident monitoring Mengesha, Isaak Owen, Branwen Collins, Charlie Wong, Tina Mylius, Simon Slattery, Peter McGregor, Sean Computers and Society Incident monitoring can drive safety improvements in high-reliability industries and population-scale technologies, but remains underdeveloped in AI governance. Public databases catalog thousands of AI incidents, but simple incident counts conflate media reporting propensity, system deployment ("exposure"), and harm frequency per unit exposure. We propose a methodological framework that accounts for these factors and calibrates confidence to available evidence in analyzing how AI incidents change over time. The framework comprises three components: a structured monitoring question that defines the scope of the analysis; a tiered estimation process that separately derives harm and exposure trends, including through LLM-assisted filtering of public incident databases; and a classification scheme that maps the resulting trend estimates onto actionable governance categories (Escalating, Mitigating, Concentrating, Receding or Unclassifiable). Through case studies, we examine the framework's clarifying power and limitations, demonstrate governance insight despite real-world data constraints, and provide a proof of concept for AI incident monitoring as a practical governance tool. |
| title | A pragmatic classification framework for AI incident monitoring |
| topic | Computers and Society |
| url | https://arxiv.org/abs/2604.21412 |