GMF

What is the GMF Taxonomy?

The Goals, Methods, and Failures (GMF) taxonomy is a failure cause analysis taxonomy interrelating the goals of the system deployment, the system's methods, and their likely failings. Details on the process are available in the recent work published for SafeAI paper.

How do I explore the taxonomy?

All taxonomies can be used to filter incident reports within the Discover Application. The taxonomy filters work similarly to how you filter products on an E-commerce website. Use the search field at the bottom of the “Classifications” tab to find the taxonomy field you would like to filter with, then click the desired value to apply the filter.

About the Responsible AI Collaborative

The AI Incident Database is a collaborative project of many people and organizations. Details on the people and organizations contributing to this particular taxonomy will appear here, while you can learn more about the Collab itself on the incident database home and about pages.

The maintainers of this taxonomy include,

Taxonomy Fields

Known AI Goal Searchable in Discover App

Definition: An AI Goal which is almost certainly pursued by the AI system referenced in the incident.

Potential AI Goal Searchable in Discover App

No classifications with this field

Definition: An AI Goal which is probably pursued by the AI system referenced in the incident.

Known AI Technology Searchable in Discover App

Definition: An AI Technology which is almost certainly a part of the implementation of the AI system referenced in the incident.

Potential AI Technology Searchable in Discover App

Definition: An AI Method / Technology which probably is a part of the implementation of the AI system referenced in the incident.

Known AI Technical Failure Searchable in Discover App

Definition: An AI Technical Failure which almost certainly contributes to the AI system failure referenced in the incident.

Potential AI Technical Failure Searchable in Discover App

Definition: An AI Technical Failure which probably contributes to the AI system failure referenced in the incident.

Known AI Goal Snippets

Definition: One or more snippets that justify the classification.

Known AI Goal Classification Discussion

Definition: Free text with comments justifying the chosen classification (e.g. based on information on selected snippets and technical analysis), if needed.

Potential AI Goal Snippets

Definition: One or more snippets that justify the classification.

Potential AI Goal Classification Discussion

Definition: Free text with comments justifying the chosen classification (e.g. based on information on selected snippets and technical analysis), if needed.

Known AI Technology Snippets

Definition: One or more snippets that justify the classification.

Known AI Technology Classification Discussion

Definition: Free text with comments justifying the chosen classification (e.g. based on information on selected snippets and technical analysis), if needed.

Potential AI Technology Snippets

Definition: One or more snippets that justify the classification.

Potential AI Technology Classification Discussion

Definition: Free text with comments justifying the chosen classification (e.g. based on information on selected snippets and technical analysis), if needed.

Snippets

Definition: One or more snippets that justify the classification.

Known AI Technical Failure Classification Discussion

Definition: Free text with comments justifying the chosen classification (e.g. based on information on selected snippets and technical analysis), if needed.

Potential AI Technical Failure Snippets

Definition: One or more snippets that justify the classification.

Potential AI Technical Failure Classification Discussion

Definition: Free text with comments justifying the chosen classification (e.g. based on information on selected snippets and technical analysis), if needed.