C1
No explanation
Reviewers see only the generated banner and make the decision themselves.
C1 : No explanation
CHI EA '26 - Published
Yonsei IRB Approved
From Workslop to Work
AI explanations do not simply calibrate trust. In creative verification, they become interface signals that decide when people stop, edit, doubt, or reclaim judgment.
Seunghyun Lim and Soojin Jun. CHI EA '26, Barcelona, Spain, April 13-17, 2026. Exploratory within-subject study with 10 marketing practitioners, 18 AI-generated banner-copy reviews, and three explanation conditions.
Participants
10 marketing practitioners, 1.5-10 years of experience
Protocol
No explanation, brief AI opinion, detailed rationale
Stress test
Trap items reveal when explanations become safety signals
The same AI explanation became assistance for some reviewers and a shortcut for others.
0.0%
of deliberately wrong AI suggestions were accepted by reviewers who treated explanations as safety signals.
0%
were accepted by reviewers who judged independently. They were also the fastest group, at 11.5 seconds per item.
same task - same explanations - opposite outcomes
Motivation
When AI drafts look complete, the unfinished work moves to the reviewer.
Workslop is not just low-quality output. It is the hidden labor of verifying, interpreting, and revising plausible but context-poor AI content. In marketing copy, reviewers cannot rely on one ground truth; they reconcile drafts with brand voice, campaign strategy, audience fit, and professional taste.
Research question
How do AI explanations reconfigure judgment labor?
Rather than asking whether explanations increase or decrease trust, the study asks how explanation granularity reshapes the concrete work of reviewing creative AI output.
Method
A trap-item study across three explanation formats
Findings 1 — Judgment Labor
Four dimensions of judgment labor
Meta-Judgment Labor
Evaluating twice over — the AI's opinion and the banner copy itself.
P5 — "Spaghetti comment. The AI throws everything at me in a lump and leaves the thinking to me."
Translation Labor
Converting a gut feeling of "something's off" into concrete, business-language critique.
P4 — "Changing the negative framing to a positive experience would fix it."
Coordination Labor
Negotiating between formal rules and marketing hooks — the practitioner's flexibility and compromise.
P8 — "It follows the general rules, but it doesn't feel attractive to consumers."
Emotional Labor
Defending one's own judgment authority against the AI's voice.
P5 — "Once a short comment sticks in my head, I have to actively hold it back."
Reviewers don't simply approve or reject — they do four overlapping forms of judgment labor while interpreting AI explanations.
Findings / response patterns
Same explanation, opposite outcomes
Judgment-Stopping
AI feedback becomes a stop signal. Reviewers accept the explanation as a safety cue and end verification early.
19.0s avg / AI match 94.5% / traps 87.5%
Editorial Intervention
AI feedback becomes a starting point for revision, helping reviewers turn a vague concern into an actionable edit.
21.5s avg / AI match 77.8% / traps 25.0%
Psychological Burden
The explanation itself becomes another object to verify, creating double work and extended uncertainty.
25.7s avg / AI match 90.7% / traps 50.0%
Critical Sovereigns
Experienced reviewers keep distance from AI feedback, use it selectively, and preserve final judgment authority.
11.5s avg / AI match 87.0% / traps 0%
Explanations are not trust-calibration tools. They are interface signals that structure the flow and responsibility of judgment.
Design implications
Three service interface hypotheses, shown as a review-tool screen.
DI 1
Risk-aware verification
Add friction only when a claim needs checking before approval.
Request for ReviewClose
Don't miss this winter - book now or you'll regret it
Unlock after checks
Limitations
Exploratory by design: N = 10, one creative domain, short-term lab setting, and pre-screened materials. The goal is analytic generalization - mapping the structure of judgment labor - not statistical effect estimation.
What's next
A follow-up study to CHI EA '26 is under IRB review with Daeun Hwang (University of Washington), Donghwan Kim, and Soojin Jun (Yonsei University).
Take-aways
Designing for judgment, not just trust.
01 / Workslop is judgment labor: a creative practice to support, not a UX bottleneck to remove.
02 / Four labor dimensions and four response patterns offer a shared vocabulary for AI-assisted creative review.
03 / Explanation design should structure timing and form, not maximize information delivered.