Workslop: Are you guilty?
Emails are getting more polished, but harder to understand. This isn't a coincidence. And it has a name.
Posted on
Mar 30, 2026
Filed under
AI & LLMs
30/3
2026
Your inbox is full. But is anyone actually saying anything?
Emails are getting longer, more polished-sounding, and harder to understand. Threads that should take two exchanges to resolve are stretching to ten. Requests arrive that sound articulate but don't actually ask for anything specific. And when people dig in to clarify, the follow-up is just as vague.
This isn't a coincidence. It's a documented, researched, and rapidly growing workplace problem. And it has a name.
What Is Workslop?
The term was coined in late 2025 by researchers from Stanford University's Social Media Lab and BetterUp Labs. They define workslop as AI-generated work content that masquerades as good work, but lacks the substance to meaningfully advance a given task.
Think: three paragraphs of polished prose that could have been one sentence. A detailed email response that doesn't actually answer the question. Slides that look impressive but contain nothing actionable. Code that runs but doesn't do what it's supposed to.
The problem isn't that people are using AI. It's that they're using it as a replacement for thinking, and then sending the output without reviewing whether it makes sense.
The Numbers Tell the Story
The Stanford-BetterUp research surveyed 1,150 full-time US employees across industries. The findings are stark:
40% of workers have received workslop in the past month.
15% of all workplace content now qualifies as workslop.
Workers spend an average of 1 hour and 56 minutes dealing with each instance, including interpreting, clarifying, correcting, or redoing the work.
That adds up to an invisible cost of roughly $186 per employee per month.
For a company of 10,000 people, that's over $9 million per year in lost productivity.
53% of recipients say receiving workslop makes them feel annoyed. 38% feel confused. 22% feel offended.
Roughly half of workers view colleagues who send workslop as less creative, less capable, and less reliable.
42% see them as less trustworthy.
32% say they're less likely to want to work with the sender again.
A separate study from Zapier found that 58% of enterprise workers spend at least three hours per week revising AI-generated content, with some logging up to 4.5 hours.
And here's the kicker: AI use at work has doubled since 2023 (from 21% to 40%, according to Gallup), yet an MIT Media Lab report found that 95% of organisations see no measurable return on their AI investment. Workslop is increasingly cited as a significant reason why.
The AI-to-AI Email Loop
The workslop research has largely focused on internal teams, specifically colleagues sending AI-generated content to each other. But there's a more damaging variant emerging in business-to-business communication that hasn't received as much attention yet.
Here's how it plays out:
One party uses ChatGPT, Gemini, or Copilot to draft an email. They haven't fully formed what they want, and they're hoping the AI will figure it out for them.
The email arrives sounding articulate and professional, but the actual request is vague, contradictory, or based on assumptions the AI made on their behalf.
The recipient uses AI to help interpret and respond, which generates a confident reply to a confused input.
The original sender receives that response, feeds it back into their AI tool, and the cycle continues.
Each pass through the loop strips away more of the original intent. The conversation drifts. Both sides sound polite and productive, but nobody is actually communicating. It's like two translation engines talking to each other: the grammar is perfect, but the meaning evaporated three emails ago.
As the BetterUp researchers put it: the insidious effect of workslop is that it shifts the burden of work downstream. In the AI email loop, that burden just keeps bouncing back and forth, growing larger each time.
Harvard Business Review covered this dynamic in March 2026, noting that when people outsource their workplace communication to AI, they're not just saving time. They're outsourcing the very moments that create connection, clarity, and shared understanding.
Why This Matters for Your Business
Whether you run a service business, manage external partners, or rely on clear communication with suppliers and collaborators, this trend has real consequences.
You waste time solving the wrong problem. If a brief or request was AI-generated and the sender didn't fully understand what they were asking for, whoever acts on it could spend hours or days delivering something that misses the mark entirely. That's not just inefficient, it damages the commercial relationship.
Trust deteriorates on both sides. The Stanford-BetterUp research found that people quickly lose respect for those who send workslop. If either side of a business relationship suspects the other is just running emails through AI without thinking, the perceived value of the relationship drops.
Clear communication becomes the exception, not the norm. When AI handles both sides of a conversation, the human skills that make communication work, such as active listening, specificity, context, and judgement, gradually atrophy. People stop thinking about what they actually need and start relying on AI to figure it out.
It's a management failure, not a technology failure. The Stanford-BetterUp researchers were direct on this point: the proliferation of workslop is a management failure. Only 19% of knowledge workers have clarity on what type of work should be done with AI, according to Asana's State of AI at Work report. When businesses encourage AI adoption without guidelines, training, or expectations around quality, workslop is the inevitable result.
How to Fix It
The good news is that this isn't an unsolvable problem. It just requires being intentional about how AI is used in communication.
Break the loop with a phone call
When an email thread starts going in circles, or when a message arrives that sounds polished but says nothing specific, pick up the phone. A five-minute call will accomplish more than ten rounds of AI-mediated email. Voice notes and short video messages (Loom, for example) are also effective at cutting through the noise.
Ask for the one-sentence version
Before actioning anything from a long, AI-flavoured email, reply with something simple: "Can you confirm in a sentence or two exactly what outcome you're after?" This forces the other person to engage their own brain rather than their AI tool.
Review before you send
If you're using AI to draft emails, which is perfectly fine, read the output as if you're the recipient. Does it actually say what you mean? Is it specific? Would someone reading it know exactly what to do next? If not, rewrite it. AI is a drafting tool, not a thinking tool.
Lead with specifics, not prose
Instead of long AI-generated paragraphs, structure your communications around concrete details: what you need, by when, and what decisions need to be made. Bullet points are underrated. Specificity is the antidote to workslop.
Set expectations within your team
Talk openly about this. Make it clear that AI-assisted communication is fine, but AI-replaced communication isn't. Set a standard: every outgoing email should pass a basic test. Does this say something a human actually thought through?
Be a "pilot," not a "passenger"
The Stanford-BetterUp research identified two types of AI users emerging in the workplace: pilots, who use AI to enhance their creativity and precision, and passengers, who rely on AI to do the work for them. Pilots review, edit, and add context. Passengers copy, paste, and send. The difference shows, and people notice.
The Bottom Line
AI is an extraordinary tool. It makes businesses faster, sharper, and more capable. But when AI becomes a substitute for thinking rather than a support for it, everyone loses.
If your emails are getting longer but your conversations are getting less clear, workslop might be the reason. The fix isn't to stop using AI. It's to start using it properly, and to have the confidence to pick up the phone when the robots start talking past each other.
Pure Agency is a Perth-based digital marketing agency. We help businesses cut through the noise with strategy, advertising, web development, and automation that actually works. If your communications could use a human touch, get in touch.
Sources:
Fortune, "Two kinds of employees are emerging in the AI-generated 'workslop' era" (October 2025)
Built In, "AI-Generated 'Workslop' Is Killing Productivity" (2025)
WebProNews, "AI Workslop: Subpar Outputs Erode Productivity and Cost Millions" (January 2026)
HR Brew, "Beyond workslop: How AI is shaping workplace communication" (February 2026)
Your inbox is full. But is anyone actually saying anything?
Emails are getting longer, more polished-sounding, and harder to understand. Threads that should take two exchanges to resolve are stretching to ten. Requests arrive that sound articulate but don't actually ask for anything specific. And when people dig in to clarify, the follow-up is just as vague.
This isn't a coincidence. It's a documented, researched, and rapidly growing workplace problem. And it has a name.
What Is Workslop?
The term was coined in late 2025 by researchers from Stanford University's Social Media Lab and BetterUp Labs. They define workslop as AI-generated work content that masquerades as good work, but lacks the substance to meaningfully advance a given task.
Think: three paragraphs of polished prose that could have been one sentence. A detailed email response that doesn't actually answer the question. Slides that look impressive but contain nothing actionable. Code that runs but doesn't do what it's supposed to.
The problem isn't that people are using AI. It's that they're using it as a replacement for thinking, and then sending the output without reviewing whether it makes sense.
The Numbers Tell the Story
The Stanford-BetterUp research surveyed 1,150 full-time US employees across industries. The findings are stark:
40% of workers have received workslop in the past month.
15% of all workplace content now qualifies as workslop.
Workers spend an average of 1 hour and 56 minutes dealing with each instance, including interpreting, clarifying, correcting, or redoing the work.
That adds up to an invisible cost of roughly $186 per employee per month.
For a company of 10,000 people, that's over $9 million per year in lost productivity.
53% of recipients say receiving workslop makes them feel annoyed. 38% feel confused. 22% feel offended.
Roughly half of workers view colleagues who send workslop as less creative, less capable, and less reliable.
42% see them as less trustworthy.
32% say they're less likely to want to work with the sender again.
A separate study from Zapier found that 58% of enterprise workers spend at least three hours per week revising AI-generated content, with some logging up to 4.5 hours.
And here's the kicker: AI use at work has doubled since 2023 (from 21% to 40%, according to Gallup), yet an MIT Media Lab report found that 95% of organisations see no measurable return on their AI investment. Workslop is increasingly cited as a significant reason why.
The AI-to-AI Email Loop
The workslop research has largely focused on internal teams, specifically colleagues sending AI-generated content to each other. But there's a more damaging variant emerging in business-to-business communication that hasn't received as much attention yet.
Here's how it plays out:
One party uses ChatGPT, Gemini, or Copilot to draft an email. They haven't fully formed what they want, and they're hoping the AI will figure it out for them.
The email arrives sounding articulate and professional, but the actual request is vague, contradictory, or based on assumptions the AI made on their behalf.
The recipient uses AI to help interpret and respond, which generates a confident reply to a confused input.
The original sender receives that response, feeds it back into their AI tool, and the cycle continues.
Each pass through the loop strips away more of the original intent. The conversation drifts. Both sides sound polite and productive, but nobody is actually communicating. It's like two translation engines talking to each other: the grammar is perfect, but the meaning evaporated three emails ago.
As the BetterUp researchers put it: the insidious effect of workslop is that it shifts the burden of work downstream. In the AI email loop, that burden just keeps bouncing back and forth, growing larger each time.
Harvard Business Review covered this dynamic in March 2026, noting that when people outsource their workplace communication to AI, they're not just saving time. They're outsourcing the very moments that create connection, clarity, and shared understanding.
Why This Matters for Your Business
Whether you run a service business, manage external partners, or rely on clear communication with suppliers and collaborators, this trend has real consequences.
You waste time solving the wrong problem. If a brief or request was AI-generated and the sender didn't fully understand what they were asking for, whoever acts on it could spend hours or days delivering something that misses the mark entirely. That's not just inefficient, it damages the commercial relationship.
Trust deteriorates on both sides. The Stanford-BetterUp research found that people quickly lose respect for those who send workslop. If either side of a business relationship suspects the other is just running emails through AI without thinking, the perceived value of the relationship drops.
Clear communication becomes the exception, not the norm. When AI handles both sides of a conversation, the human skills that make communication work, such as active listening, specificity, context, and judgement, gradually atrophy. People stop thinking about what they actually need and start relying on AI to figure it out.
It's a management failure, not a technology failure. The Stanford-BetterUp researchers were direct on this point: the proliferation of workslop is a management failure. Only 19% of knowledge workers have clarity on what type of work should be done with AI, according to Asana's State of AI at Work report. When businesses encourage AI adoption without guidelines, training, or expectations around quality, workslop is the inevitable result.
How to Fix It
The good news is that this isn't an unsolvable problem. It just requires being intentional about how AI is used in communication.
Break the loop with a phone call
When an email thread starts going in circles, or when a message arrives that sounds polished but says nothing specific, pick up the phone. A five-minute call will accomplish more than ten rounds of AI-mediated email. Voice notes and short video messages (Loom, for example) are also effective at cutting through the noise.
Ask for the one-sentence version
Before actioning anything from a long, AI-flavoured email, reply with something simple: "Can you confirm in a sentence or two exactly what outcome you're after?" This forces the other person to engage their own brain rather than their AI tool.
Review before you send
If you're using AI to draft emails, which is perfectly fine, read the output as if you're the recipient. Does it actually say what you mean? Is it specific? Would someone reading it know exactly what to do next? If not, rewrite it. AI is a drafting tool, not a thinking tool.
Lead with specifics, not prose
Instead of long AI-generated paragraphs, structure your communications around concrete details: what you need, by when, and what decisions need to be made. Bullet points are underrated. Specificity is the antidote to workslop.
Set expectations within your team
Talk openly about this. Make it clear that AI-assisted communication is fine, but AI-replaced communication isn't. Set a standard: every outgoing email should pass a basic test. Does this say something a human actually thought through?
Be a "pilot," not a "passenger"
The Stanford-BetterUp research identified two types of AI users emerging in the workplace: pilots, who use AI to enhance their creativity and precision, and passengers, who rely on AI to do the work for them. Pilots review, edit, and add context. Passengers copy, paste, and send. The difference shows, and people notice.
The Bottom Line
AI is an extraordinary tool. It makes businesses faster, sharper, and more capable. But when AI becomes a substitute for thinking rather than a support for it, everyone loses.
If your emails are getting longer but your conversations are getting less clear, workslop might be the reason. The fix isn't to stop using AI. It's to start using it properly, and to have the confidence to pick up the phone when the robots start talking past each other.
Pure Agency is a Perth-based digital marketing agency. We help businesses cut through the noise with strategy, advertising, web development, and automation that actually works. If your communications could use a human touch, get in touch.
Sources:
Fortune, "Two kinds of employees are emerging in the AI-generated 'workslop' era" (October 2025)
Built In, "AI-Generated 'Workslop' Is Killing Productivity" (2025)
WebProNews, "AI Workslop: Subpar Outputs Erode Productivity and Cost Millions" (January 2026)
HR Brew, "Beyond workslop: How AI is shaping workplace communication" (February 2026)
Author
Steven Donald
Chief Strategist
With over 30 years of experience across all facets of digital marketing, Steven Donald brings this expertise to his role as Chief Strategist at Pure Agency. Having navigated every evolution from early digital transformation to today's AI-driven landscape, Steven possesses a unique perspective on what truly drives performance.
