---
id: "concept-workslop-d79"
type: "concept"
source_timestamps: ["¶3"]
source_url: "https://hbr.org/2026/02/how-to-foster-psychological-safety-when-ai-erodes-trust-on-your-team"
source_title: "How to Foster Psychological Safety When AI Erodes Trust on Your Team"
tags: ["productivity", "ai-output", "team-friction"]
related: ["concept-trust-ambiguity", "claim-ai-errors-ripple-differently"]
definition: "AI-generated output that fails to advance a project and creates extra cognitive and emotional labor for colleagues who must fix it."
sources: ["adoption"]
sourceVaultSlug: "hbr-seg-adoption"
originDay: 9
articleStem: "hbr-cl-79-psychological-safety-ai-trust"
sourceUrl: "https://hbr.org/2026/02/how-to-foster-psychological-safety-when-ai-erodes-trust-on-your-team"
sourceTitle: "How to Foster Psychological Safety When AI Erodes Trust on Your Team"
---
# Workslop

**Workslop** is AI-generated output that *fails to advance a project* and instead dumps a burden of extra cognitive and emotional labor onto the human colleagues who are forced to fix, verify, or redo the work. The authors introduce the term (¶3) to name a now-common failure mode: tools that were ostensibly deployed to make people's jobs *easier* end up generating messes that coworkers must clean up.

The damage is twofold. First, workslop **directly harms organizational productivity** — the promised time savings are clawed back (and then some) by rework. Second, and more insidiously, it **damages interpersonal trust among coworkers**, because the person who ships unreviewed AI output effectively offloads their labor onto teammates. This links workslop tightly to [[concept-trust-ambiguity]] (people stop trusting both the tool and each other) and to [[claim-ai-errors-ripple-differently]] (the rework cannot be metabolized through the normal human error-recovery process).

**External grounding:** The *label* "workslop" appears to be coined in this article, but the underlying phenomenon is well documented. Nature's study on the "dark side of AI adoption" notes that AI creates extra stressors — monitoring, uncertainty, error-checking — that raise cognitive load; TechUK and practitioner sources (e.g., Seth Mattison) similarly describe the burden of verifying AI outputs and handling hallucinations as a real source of lost productivity and frustration.


## Related across articles
- [[concept-workslop-d38]]
- [[concept-workslop-d42]]