---
id: "question-defining-quality-ai"
type: "open-question"
source_timestamps: ["§ The Workslop Pressure Cooker"]
tags: ["metrics", "quality-assurance"]
related: ["claim-blanket-mandates-fail", "lit-digital-taylorism"]
resolutionPath: "Developing role-specific rubrics and KPIs that measure the actual business value and accuracy of AI-assisted tasks, rather than just tracking the frequency of tool usage."
sources: ["adoption"]
sourceVaultSlug: "hbr-seg-adoption"
originDay: 9
articleStem: "hbr-edu-38-ai-workslop"
sourceUrl: "https://hbr.org/2026/01/why-people-create-ai-workslop-and-how-to-stop-it"
sourceTitle: "Why People Create AI “Workslop”—and How to Stop It"
---
# How do organizations define 'quality' AI output?

The article highlights that executives prize 'quantity and use of AI' over 'quality and effectiveness,' and that there is 'no specification on what quality AI output looks like specific to work.' The text does not provide a universal metric, leaving it a gap for organizations to solve. It is tightly bound to [[claim-blanket-mandates-fail]] and echoes the [[lit-digital-taylorism|fake-work]] critique of activity-over-impact metrics.

**Resolution path:** Develop role-specific rubrics and KPIs that measure the actual business value and accuracy of AI-assisted tasks, rather than the frequency of tool usage.