Attribution
A stable, linkable reference for describing how AI contributed to a piece of work. The goal is a simple, readable signal about human-AI involvement, not a questionnaire, not a disclaimer.
This page defines the taxonomy itself as a public standard. The badges used across this site are one implementation of that standard in publishing practice.
| Level | Name | Meaning |
|---|---|---|
| AI-A | Assisted | AI provided suggestions, autocomplete, or lookup; human wrote and decided everything. |
| AI-E | Enhanced | AI drafted or transformed content; human substantially revised, curated, and owns the result. |
| AI-C | Collaborative | Human and AI iterated together; human directed, reviewed, and takes responsibility. |
| AI-G | Generated | AI produced the primary content; human reviewed and chose to publish. |
Further reading
Full trail →Most AI acknowledgments are too vague to be useful. Process transparency gives teams a practical, auditable way to describe human-AI work without pretending the model is an author.
Current AI attribution approaches fail to address real-world adoption challenges. While industry solutions like IBM's toolkit offer important first steps, they miss the social and professional dynamics that shape practice across different professional contexts. Drawing from my experience integrating GenAI across multiple courses and developing systematic attribution practices, here's what's missing from current research.
An independently developed framework proposing standardized transparency protocols for human-AI collaboration across journalism, academic, and creative professional domains. Developing practical framework with four attribution categories positioned for adoption similar to Creative Commons licensing model.