Worklop epidemic or how AI kills trust in you

You’ve surely encountered this. Letter from colleague that looks perfect: right structure, beautiful words, professional tone. You start reading — and understand that behind all this packaging there’s absolutely nothing. No specifics, no solutions, just beautifully packaged emptiness. Congratulations: you just encountered workslop.

Workslop is new word describing old problem, amplified by AI to catastrophic proportions. Fresh study from Stanford University and BetterUp Labs showed such statistics: 53% of employees experience irritation when receiving such content from colleagues.

And for company with 10,000 employees this costs $9 million in productivity losses annually. $9 million! And these are just numbers — we’re not yet talking about destroyed trust, ruined relationships and burned-out employees.

So, what is workslop? Kate Niederhoffer, Vice President of BetterUp Labs and social psychologist, together with colleagues from Stanford gave clear definition in Harvard Business Review. “Workslop is when employees use AI tools to create work requiring minimal effort and looking acceptable. But ultimately creating more work for their colleagues”. Term workslop appeared by analogy with “AI slop” — meaningless AI content that flooded internet. Simply put, this is when your colleague pressed “generate” button, copied result and sent it to you. Without even bothering to check if it makes any sense.

Jeffrey Hancock, communication professor at Stanford and founder of Stanford Social Media Lab, gives vivid example from study: “Receiving this low-quality work created enormous time loss and inconvenience for me,” told one project manager. “Since this was provided by my supervisor, I felt awkward. Confronting her about low quality and asking to redo. So instead I had to take on effort to do what should have been her responsibility, which hindered my other ongoing projects”.

Now most interesting: who creates all this workslop?

According to Harvard Business Review study, 40% of such content comes from colleagues at your level. Another 16% — from management. Yes-yes, your boss can also be workslop generator!

Why do even managers fall into this trap? Jeffrey Hancock explains. Without guidance and leadership, workers can act from fear. If they don’t use AI, they’ll be replaced, but if they use it, they’ll be judged.

What does workslop cost? As research team writes in Harvard Business Review: “Employees report spending average of 1 hour 56 minutes sorting through each workslop case”. Almost 2 hours! Niederhoffer and her colleagues calculated: “We base on participants’ estimates of time spent as well as their stated salary. And we found that these workslop incidents carry invisible tax of $186 per month. For organization of 10,000 workers, this yields over $9 million per year of lost productivity. Given estimated workslop prevalence of 41%”.

But price of workslop isn’t just money. This is destroyed trust. Kate Niederhoffer emphasizes important psychological aspect: “Workslop phenomenon is connected with shifting burden to another person without being aware of this impact. People forget this because we think of AI as tool we alone work with. But actually it mediates work between people”.

Numbers are terrible: 42% of workslop recipients start trusting sender less. 37% consider them less intelligent. 54% see them as less creative.

Emotional reaction to workslop is also impressive. Study shows: 53% of people experience irritation when encountering low-quality AI-generated content, 38% feel confusion, 22% say such behavior from colleagues offends them.

Now about career risks. As study authors note: “32% of people who received workslop report they’re less likely to want to work with sender again in future”. Almost third! In world where success often depends on contact network and reputation, this is deadly blow. You can be brilliant specialist, but if you’re perceived as person who sends useless garbage, doors will close.

So why do people create workslop? Paradoxically, main reason — desire to save time. Niederhoffer explains problem thus: “Previously effort was required to create content. Now, when this element of effort has disappeared, I can generate lots of useless or unproductive content very easily”. This is productivity illusion — you create more, but not better. Quantity without quality.

Connection between workslop and general AI inefficiency in business is obvious. MIT report “The GenAI Divide: State of AI in Business” showed that 95% of pilot projects with generative AI don’t achieve rapid revenue growth. Aditya Challapalli, lead author of report and researcher at MIT Media Lab, explains problem: “It’s not about quality of AI models, but about training gap for both tools and organizations”. Report emphasizes that unlike consumer tools like ChatGPT, enterprise AI requires careful integration into existing systems.

How to avoid workslop? Jeffrey Hancock gives simple but important advice: “What reduces workslop — is team’s commitment to task quality”.

He also recommends companies focus on organized approach to implementing and promoting AI at work. If you use AI, inform colleagues. As Hancock notes: “If you tell your colleague that work you’re sending is AI-created, they’ll be able to better understand what prompts you worked with. And what your goal was, and fill any gaps”.

Kate Niederhoffer advises leaders to focus on human agency and encourage “pilot thinking”. To see how tools can give more control in workplace. She emphasizes: “High agency with AI can be incredible. But this is in sharp contrast to simple copy-paste mode, when you just let tool do all work for you. And forget to let it complement your human competencies”.

Before sending something created with AI, ask yourself 3 questions. First: do I myself understand what’s written here? If no — don’t send. Second: does this really answer question or solve task? If no — don’t send. Third: would I be proud of this work if I wrote it myself? If no — you know answer.

Workslop isn’t just annoying trifle. This is indicator of how we relate to new technologies, to our work and to each other. When you send unchecked AI text, you’re not saving time — you’re stealing it from colleagues. You’re not increasing productivity — you’re killing it. And most importantly — you lose what’s impossible to restore with one mouse click: trust and respect of people you work with.

AI is incredible tool. But tool requires master, not someone who just presses buttons at random. Next time when you want to press “generate” and immediately “send”, remember this number: almost 2 hours of someone else’s time wasted. Remember that third of people won’t want to work with you anymore. Remember $9 million in losses. And ask yourself: was it worth it?