← The Calibrated Authority Index
university

University College London

9/12Calibrated Authority
“The student should still be the author of their own work – in other words, GenAI should be limited to supporting and assisting the student in completing the assessment, and should not be completing the assessment (entirely or only in part) on the student’s behalf.”— UCL Teaching & Learning, Generative AI Hub, Three categories of GenAI use in assessment, Category 2 (read via Wayback capture 2026-07-22) · source ↗
D1
Traceability & inspectability
1
D2
Human authorship & accountability
2
D3
Disclosure & labeling
2
D4
Synthetic-identity / fabrication prohibition
1
D5
Human validation in loop
2
D6
Evidential-trust emphasis
1
Posture Enabling Boundary fit · fits Trust-logic Both-split ⚑ twilight framing Updated 2026-09-27 · rev 1
Source policy: https://www.ucl.ac.uk/teaching-learning/generative-ai-hub/three-categories-genai-use-assessment
Machine record: /institutions/ucl.json

Cite this

Reitz, C.H. (2026). The Calibrated Authority Index — University College London. https://calibrated-authority.chrishuberreitz.com/institutions/ucl Quote: “The student should still be the author of their own work – in other words, GenAI should be limited to supporting and assisting the student in completing the assessment, and should not be completing the assessment (entirely or only in part) on the student’s behalf.” Source: https://www.ucl.ac.uk/teaching-learning/generative-ai-hub/three-categories-genai-use-assessment
Index Methodology Manifest JSON atom CC-BY-4.0 · v2026-11-09 · Chris Huber Reitz