Companies Are Losing Money To AI 'Workslop' That Slows Everything Down

Modern workplaces, seeking the promise of innovation, speed, and efficiency, have rapidly adopted artificial intelligence. However, this quick adoption often results in a "messier" reality, as companies pressure themselves to integrate AI quickly to avoid falling behind competitors.

New research indicates that rather than saving time, poorly managed AI implementation is actively costing large organizations millions of dollars and thousands of man-hours annually.

This drain on resources is due to a phenomenon termed "workslop" by the Harvard Business Review (HBR).

While large language models (LLMs) are adept at producing grammatically correct sentences, their output frequently lacks clarity and accuracy. Without necessary human review, these AI-generated documents create more confusion than progress, shifting effort "downstream" and bogging down the very workplace processes AI is meant to make faster.

Research conducted by HBR's BetterUp Labs and the Stanford Social Media Lab highlights the scale of this problem.

These studies show that AI-generated documents often appear "polished" but lack the required substance needed to advance a task.

A Stanford ongoing survey of US-based full-time employees found that a significant 40 percent reported receiving such low-quality outputs in the past month.

The time expenditure for cleanup is considerable. Workers spend nearly two hours per incident correcting or interpreting these substandard outputs. When multiplied across large companies, these lost hours result in thousands of lost workdays each year and millions of dollars in wasted effort, representing a significant hidden cost.

Despite the challenges, AI adoption is soaring, with Gallup reporting that the number of US employees using AI at least a few times annually has nearly doubled in recent years.

Yet, this increasing usage is often not yielding returns. An MIT Media Lab study found that fewer than one in ten AI projects delivered real revenue gains, cautioning that "95 percent of organizations are getting zero return" on their AI investments.

Beyond the financial impact, the social and emotional cost of dealing with workslop is substantial.

The frustration caused by failed automation shows clearly in employee accounts. One retail director, cited by the HBR, described his deep dissatisfaction with the company’s AI rollout.

He explained: "I had to waste more time following up on the information and checking it with my own research." He added that he then spent even more time arranging meetings with other supervisors and, in the end, had to redo the work himself.

This frustration is widespread. Survey results reveal significant emotional consequences for recipients of AI workslop:

• Over half of respondents (53 percent) reported feeling annoyed after receiving low-quality AI outputs.
• Nearly a quarter (22 percent) reported feeling offended.

Furthermore, colleagues who submit low-quality AI work are often perceived as less reliable or capable, demonstrating how AI missteps can "ripple through team dynamics".

Experts point out that the main issue is not the technology itself but the way companies choose to roll it out.

Broad mandates to use AI in every situation often result in careless copy-paste usage rather than careful, targeted application. To reduce workslop, researchers suggest three key practices:

• Set Clear Boundaries: Define exactly where AI is genuinely useful.
• Design Thoughtful Workflows: Put in place structured processes for integrating AI.
• Lead by Example: Managers should demonstrate selective and purposeful use of AI.

In practice, this may mean restricting AI to early drafts or routine summaries. Final outputs must still undergo human review. When leaders model this selective and careful use, employees are more likely to see AI as a helpful tool instead of a shortcut.