Approximately 40% [1] of U.S. workers said they have received "workslop," which is low-quality AI-generated content that appears polished but lacks substance.
This trend is creating a productivity paradox where tools designed to save time instead shift the burden of labor to the recipient. Because AI can produce text that looks professional while containing factual errors or poor reasoning, employees must spend significant time auditing and rewriting the material.
Alexi Robichaux, CEO of BetterUp, said that fixing a single piece of workslop can take around two hours [1]. This process involves stripping away the superficial polish to correct the underlying logic and accuracy of the document.
The phenomenon occurs because large language models are designed to be persuasive and fluent rather than strictly accurate. This creates a gap between the perceived quality of a document and its actual utility, a gap that often goes unnoticed until the recipient attempts to use the information.
Robichaux said that this cycle reduces overall efficiency. When a sender uses AI to generate a report in seconds, they may believe they have increased productivity, but they have instead offloaded the effort of quality control to the person receiving the file.
Addressing workslop requires a shift in how organizations view AI output. Relying on the surface-level appearance of a document can lead to systemic errors if the underlying reasoning is flawed. The time spent correcting these errors often outweighs the time saved during the initial generation process [1].
“Fixing a single piece of workslop can take around two hours.”
The rise of workslop suggests that the initial wave of AI-driven productivity gains may be illusory. As the cost of generating content drops to near zero, the cost of verification increases, potentially creating a new form of corporate inefficiency where human workers spend more time editing AI hallucinations than they would have spent writing from scratch.
