The Hidden Costs of AI-Generated Work: Can Businesses Recover Lost Productivity?
The promise of artificial intelligence (AI) in the workplace was always clear: automate routine tasks, unleash creativity, and boost productivity across industries. Yet a surge of new research suggests the reality is far more complex—and quietly expensive. Stanford University, in collaboration with BetterUp Labs, has identified a troubling paradox: the widespread use of generative AI has led to an epidemic of “workslop”—polished but substantively weak output that passes as good work but requires significant human correction.
What Is AI Workslop—and Why Is It Expensive?
The term “workslop” describes AI-generated work that looks convincing but fails to advance meaningful business objectives. According to the Stanford-BetterUp study published in Harvard Business Review, over 40% of employees reported encountering workslop in the past month. Each instance costs colleagues almost two hours to interpret, fix, or redo, resulting in a hidden “tax” of $186 per employee per month. For a mid-sized organization of 10,000, this translates to $9 million annually in lost productivity.
Unlike traditional automation, which aims to reduce cognitive labor, workslop actually offloads the burden onto others downstream. The frustration doesn’t just waste time—it undermines trust and morale across teams.
The Toll on Workplace Relationships
The fallout from workslop is more than just financial. How do colleagues react to receiving it? Surveys reveal that 53% feel annoyed, 38% confused, and 22% outright offended. Half of recipients judge senders as less creative, capable, and reliable, while over one-third consider them less trustworthy or intelligent. Professionals in technology and services industries are especially affected, raising workplace tensions and threatening future collaboration.
Direct quotes from the research highlight the ripple effect:
“I had to waste more time checking information and setting up meetings to redo the work myself.”
“It created a situation where I had to decide whether to redo the work or just call it good enough.”
The Productivity Paradox: Why Aren’t Companies Seeing AI ROI?
Despite billions in enterprise spending on generative AI—$30-40 billion last year alone—95% of organizations report no measurable business return. MIT Media Lab’s recent assessment reinforces the paradox: most companies are adopting AI but aren’t getting transformational results, largely stuck in experimentation without the process changes needed for success.
Stanford-BetterUp researchers note the difference between “pilots” (who use AI purposefully to enhance creativity) and “passengers” (who use it as a shortcut to avoid work). The latter group may embrace AI without guidance or quality standards, leading to a rise in low-effort, high-volume output—workslop.
Solutions: How Leaders Can Fight Workslop
To counteract the hidden costs of AI-generated work, companies need to:
Model purposeful, thoughtful AI use at the executive level.
Set clear guidelines and norms for when and how AI should be used.
Measure performance and feedback against human-AI standards as rigorous as for traditional work.
Empowering employees to view AI as a collaborative partner—rather than a replacement for effort—may help realize the creative and productive potential envisioned for the technology.
Final Thoughts
The research underscores a pivotal truth: the journey to optimizing AI in the workplace is a marathon, not a sprint. The initial dip in productivity may be inevitable, but organizations mature in digital transformation tend to bounce back stronger. For now, recognizing and rooting out workslop is essential for companies aiming to recover trust, protect collaboration, and secure real value from their AI investments