Taxonomy
A system for organizing and categorizing information, often in a hierarchical structure.
A taxonomy is a way to organize and classify things. Think of it like a library's system for sorting books by genre, author, and subject. This helps people find what they are looking for more easily.
In AI, taxonomies are used to categorize data, identify different types of content, or group similar concepts. For example, a system might use a taxonomy to classify images into categories like 'animals', 'vehicles', and 'landscapes'. This helps the AI understand and process information more effectively.
A taxonomy is a structured classification system, typically hierarchical, used to organize concepts, data, or entities. It defines relationships between categories, such as parent-child or sibling relationships, enabling systematic organization and retrieval. Taxonomies are fundamental in areas like knowledge representation, data labeling, and content moderation, providing a framework for understanding and managing complex information spaces.
In the context of AI safety, taxonomies are crucial for defining and detecting undesirable content. A well-defined taxonomy allows for granular classification of harmful or inappropriate material, aiding in the development of robust detection models and policy enforcement.
Recent discussions around managing AI risks and improving developer tools highlight the need for clear categorization. Microsoft's work on open-weight model risk and GitHub Copilot's features for defining governance rules both rely on organizing information effectively. OpenAI's approach to content detection also involves classifying different types of undesirable outputs.
- Managing open-weight model risk in enterprise deploymentsMicrosoft · Microsoft Azure Blog ·
- GitHub Copilot suggests custom properties definitionsMicrosoft · GitHub Changelog: Copilot ·
- A Holistic Approach to Undesired Content Detection in the Real WorldOpenAI · OpenAI News ·
- Microsoft Copilot
- Microsoft uses taxonomy to classify AI-related risks in enterprise deployments and to organize property definitions within GitHub Copilot.
- OpenAI ChatGPT
- OpenAI employs a taxonomy as part of its system for detecting undesired content in AI applications.
- Ontology
- An ontology is a more formal and expressive representation of knowledg
- Classification
- Classification is the process of assigning an item to one or mor
- Knowledge Graph
- A knowledge graph represents entities and their relationships i
The term 'taxonomy' originates from biology, specifically from the work of Carl Linnaeus in the 18th century for classifying living organisms. The concept has since been adopted and adapted across many scientific and information management fields. Source ↗
Ask an AI chatbot to list different types of dogs and their breeds. Observe how it organizes the information, likely using a hierarchical structure similar to a taxonomy. You can then ask it to explain the relationship between different breeds or categories.
The Playbook: reusable how-tos, prompts, configs and tips pulled out of the news.
- HeuristicHolistic approach to content moderation systemsBuild robust content moderation with a chain of carefully designed steps. · 120 min
- PromptRole-play a student struggling with math for AI tutor simulationSimulate a student's math struggles and uncertainty for AI tutor evaluation. · 5 min
- TemplateReusable workflow templates for healthcare teamsTeams can create and share templates for common workflows like documentation or prior authorizations. · 15 min
- TemplateTemplate for building an AI tool tailored to your workA structured approach to defining and building a custom AI tool for specific business needs. · 15 min
- TemplateEvals CSV prompt list for agent skillsA CSV file template for listing prompts used in evaluating agent skills. · 20 min
Questions people ask
- What is the difference between a taxonomy and a list?
- A list is a simple collection of items, while a taxonomy organizes items into categories and often shows relationships between them, like parent-child structures.
- How are taxonomies used in AI?
- Taxonomies help AI systems understand, organize, and categorize data, which is essential for tasks like content moderation, search, and recommendation systems.
First seen on Subvolts Jun 20, 2024. Updated Oct 18, 2026. Written by an AI model from the items above and reviewed against them weekly; report an error.