More Human or More AI? Visualizing Human-AI Collaboration Disclosures in Journalistic News Production

More Human or More AI? Visualizing Human-AI Collaboration Disclosures in Journalistic News Production

Source: cs.CY updates on arXiv.org — https://arxiv.org/abs/2601.11072 Date read: 2026-06-13 Connected to: none Escalation: store-only Escalation rationale:

What this is

A design-research paper (co-design + prototype + lab study) exploring visual disclosure formats for human-AI collaboration in news production. The work generates design artifacts and tests reader comprehension/trust across four visualization approaches and two collaboration ratio conditions.

What I took from it

This is primarily a tool/UX design contribution rather than a theoretical or empirical claim about laws governing protocolized systems. The paper's value lies in identifying that existing disclosure labels ("AI-assisted," etc.) fail to communicate structure of collaboration—specifically task distribution, temporal sequencing, and agency allocation. However, the work does not propose or test a mechanism explaining why certain disclosure formats produce better outcomes, nor does it generalize beyond journalistic contexts.

The study finds that visualization type and collaboration ratio both affect reader perception, but this is a domain-specific interaction effect rather than a claim about how transparency protocols function across artificial systems. The paper is instrumentally useful for journalism design but does not contribute evidence toward laws of protocolized systems or extend active hypotheses about transparency, legitimacy, or system behavior.

Research connections

• none directly applicable

Candidate laws or signals

CL-2601.11072-1: Structural transparency beats label transparency in human-AI systems. Readers better calibrate expectations when given task-role-timeline information vs. categorical labels—but this may be specific to cognitive tasks where collaboration choreography is salient.