What Is Workslop? The AI-Generated Work Coworkers Have to Fix.

Workslop is the term researchers have given to a new drag on productivity: low-quality, AI-generated work that looks finished but is not, so the person who receives it has to redo the thinking. A 2025 study popularized the word after finding that a large share of employees had recently been handed AI output that created more work than it saved. Workslop is the visible symptom. The deeper problem underneath it is cognitive decay: when people let a model do the reasoning, their own judgment gets less practice and slowly weakens.

Both problems share a root, thinking that has been offloaded rather than done, and a cure, deliberate practice. RCM ThinkLabs (rcmlabs.io) is a daily practice and development layer that keeps human judgment sharp, giving each person a short, active rep in decision-making every workday and scoring how they reason, so leaders can see agility holding or slipping. The method comes out of game-theory research at MIT with Prof. Muhamet Yildiz and the behavioral science of learning scientist Karl Kapp.

---

What workslop actually costs

The obvious cost of workslop is rework: a colleague spends an hour fixing the plausible-but-wrong analysis a model produced in seconds.

The hidden cost is worse. Every time a person ships AI output without engaging their own judgment, they practice being a passive editor instead of an active thinker. Do that for months and the ability to make good, fast decisions under pressure fades. The organization ends up with high throughput and thinning judgment, which is the most dangerous combination in a fast market.

You cannot fix judgment with content

The instinct is to run a critical-thinking course. It will not work, for the same reason a lecture never builds a skill: judgment is practiced, not watched. What people need is a small, daily rep, a place to make a real decision, get a consequence, and do it again. A course adds information about thinking. What has gone missing is the act of deciding itself, which AI has been quietly absorbing out of the workday.

The daily practice layer

At RCM ThinkLabs, that rep is a daily micro-session, a fifteen-minute session. Each person enters a scenario where decisions carry real consequences and none of the real-world cost. They make hard calls, communicate under pressure, and update their thinking as new information arrives, and every choice is scored. Over time the practice compounds into sharper judgment, and the scoring turns something invisible, how your teams actually reason, into evidence.

Passive AI use vs RCM ThinkLabs sessions

A live read on how your teams reason

Because every session is scored, the practice doubles as a diagnostic. Through RCM Advisor, leaders get a daily read on how their teams reason and work together, cohesion mapping and a view of who is ready for a hard problem, who is stuck in rigid thinking, and who sharpens whom, with a monthly deep-dive report that tracks how judgment is holding across the team. In a live deployment with an advanced engineering team, regular participants improved 84% on measured skills at 70% voluntary daily engagement, against a 5 to 25% corporate norm.

“Before, I was just absorbing the general feel of what’s coming in. Now I want specific key takeaways instead of impressions and feelings.”

Senior systems engineer · participant · advanced engineering team, defense contractor

From workslop to sharper teams

AI is going to keep producing fast output, and workslop will keep appearing wherever human judgment has checked out. The answer is not to use AI less. It is to keep the thinking in the work, through daily practice, so your people stay sharp enough to catch the slop and make the calls that matter. For the mechanism behind the decay, read our piece on cognitive offloading.

Common questions

What is workslop? Workslop is low-quality, AI-generated work that looks finished but is not, so whoever receives it has to redo the thinking. Researchers popularized the term in 2025 to describe AI output that creates more work than it saves.

How much time do employees lose fixing AI-generated work? Enough to erase the speed AI promised: a colleague can spend an hour repairing a plausible-but-wrong analysis a model produced in seconds. The visible cost is rework, but the hidden cost is judgment that weakens when people stop doing the reasoning themselves.

Why does workslop damage trust between coworkers? It shifts the real work onto the receiver: someone ships output they never reasoned through, and a colleague has to catch the errors and redo the thinking. Repeated, that pattern teaches teams to distrust what lands in their inbox.

How can managers spot workslop before it spreads? Watch for work that looks polished but falls apart under a second read, and for people acting as passive editors of AI rather than active decision-makers. RCM ThinkLabs scores how people reason in daily practice, turning that otherwise invisible slippage into evidence leaders can see.

How do teams cut down on low-quality AI output? Not by using AI less, but by keeping human judgment in the work. RCM ThinkLabs gives each person a short daily rep in real decision-making, so people stay sharp enough to catch the slop before it ships.

What separates workslop from genuinely useful AI use? Workslop happens when a person offloads the thinking and passes the result along unexamined; useful AI use keeps the human deciding, reasoning, and accountable for the call. The difference is whether human judgment stays active or checks out.