Transparent AI Use

Transparent AI use is the practice of making AI participation visible enough that a teacher, collaborator, or reviewer can judge what the human did and what the tool contributed. In Teaching students to ‘be better than a robot’, Christy Gerdhary describes a writing assignment where students collaborate with AI and color-code which material is theirs and which belongs to the bot.

The concept differs from generic disclosure because it is process-oriented. The student is not only saying “AI was used”; the student is showing the boundary between prior writing, prompt-driven transformation, AI-generated material, human revision, and final meaning. That makes transparency a learning tool as well as an integrity tool.

Transparent AI use is especially important where AI Detector Bias makes automated suspicion risky. It gives educators a constructive alternative: ask students to make collaboration inspectable and reflective rather than depending only on after-the-fact detection.

Key Claims

• Transparency should reveal process, authorship, and judgment, not only the presence or absence of a tool.
• Color-coding or annotation can turn AI collaboration into evidence of learning rather than evidence of cheating.
• Transparent use lets students compare generated transformations against their own goals and voice.
• The method works best when paired with assignments that require personal context, embodied work, or explanation of choices.
• Transparency can reduce reliance on AI detectors, but it still depends on clear classroom norms and teacher judgment.

Connections

• AI Writing Pedagogy - classroom frame where this concept is introduced.
• Christy Gerdhary and The Generator - source speaker and faculty group.
• AI Detector Bias - reason transparent process can be fairer than detector-first policing.
• AI Content Provenance - adjacent provenance concept for synthetic media and source visibility.
• Human Judgment Under AI, First Draft Thinking, and Learning Experience Design - broader judgment, timing, and assignment-design context.