AI Work Slop Is a Mirror

Read time: 8 minutes | Key metric: £7.4M annual cost | Action required: Immediate organisational restructuring

The latest de-cry of AI is the work slop employees are generating on the daily, those that have access to AI chat bots anyway.

Your employees are using AI to produce industrial quantities of corporate trash because that’s exactly what you’ve been rewarding them for doing manually since 1987. The slop isn’t the disease—it’s your organisation finally vomiting up decades of fake work. Cost of diagnosis? £740 per employee per month. Cost of ignoring it? Your relevance.

AI work slop is going to cost the UK enterprises £7.4M annually for a 10,000-person company. But here’s the real cost: it’s revealing your company runs on performance theatre, not performance. The first movers who fix this will capture disproportionate value. The rest? They become case studies at business schools, filed under “Digital Darwinism: Companies That Blamed the Tools.”

The Situation Room

Let’s get clear on what’s actually happening here, because the narrative you’re being sold is as useful as a chocolate teapot.

The Myth: “AI is producing garbage content that’s destroying productivity.” The subtext whispers: technology is the villain, we’re the victims, someone should really do something about this terrible AI making all this terrible content.

The Reality: AI is a mirror reflecting your broken incentive structure back at you in 4K resolution. That Stanford/BetterUp study showing 40% of workers receiving work slop? They missed the punchline—it’s because 80% of assigned work has always been slop-worthy. AI just gave it a velocity upgrade.

The Stakes: Companies fixing this gain 47% productivity. Those blaming AI? They’re losing talent faster than Twitter lost advertisers. McKinsey buried the lede in their latest report: organisations addressing root causes see 3.2x ROI versus those just buying “better AI.” That’s the difference between transformation and expensive procrastination.

As Ethan Mollick puts it with characteristic understatement: “We’ve been teaching AI to reproduce our worst workplace habits. It learned perfectly.”

The Evidence Cascade

The PowerPoint Industrial Complex

Your company spends £2.3M annually on presentations nobody reads. This isn’t hyperbole—this is math.

The average enterprise creates 30,000 PowerPoints annually. Microsoft’s internal data—yes, they tracked their own digital waste—shows 87% are never opened after the initial meeting. Each deck averages 47 slides. Readers retain 4% of the content. That’s 1.88 slides of retention per presentation, if you’re keeping score. At an average creation time of 3.4 hours per deck, you’re burning 102,000 human hours annually on digital landfill.

Now AI can produce these monuments to meaninglessness in 10 minutes. Same worthlessness, 95% time savings. It’s like innovating a faster way to dig holes and fill them back up.

Remember Enron’s 500-page risk reports? They were pristine. They were comprehensive. They contained every conceivable metric, projection, and analysis. They were also fiction, wrapped in Excel, delivered via PowerPoint. AI just democratised Enron-level documentation theatre. Now everyone can produce board-ready bollocks at scale.

Which means your AI isn’t broken—it’s perfectly reproducing the meaningless work you’ve been requiring for decades. It’s a master forger, and you’ve been running a counterfeit factory.

The Productivity Paradox Proof

Here’s a number that should keep you up at night: Companies mandating AI use without restructuring see negative 12% productivity. Not a typo. Negative.

Google’s study of 5,000 developers tells the whole story. Code output up 41%. Code instability up 38%. It’s like measuring a surgeon by incisions per hour—impressive until you count the corpses. These developers are now spending 2.4 hours daily “managing AI output,” which is corporate speak for “fixing the mess our tools created because we asked for the wrong thing.”

Cost per developer in “AI overhead”? £47,000 annually. That’s not the AI subscription—that’s the human cost of managing AI that’s been deployed into a broken system. You’re paying humans to babysit robots that are perfectly executing terrible instructions.

It’s like giving everyone Ferraris but keeping the speed limit at 20mph and wondering why your fuel costs skyrocketed while arrival times stayed the same. Impressive capability, worthless outcomes, expensive confusion.

The Great Resignation 2.0 Signal

Your best people are going to leave over work slop. Not because AI will replace them—because AI revealed their jobs were meaningless all along.

LinkedIn data shows “work with purpose” searches up 340% among AI-exposed roles. But here’s the twist Shyamalan would appreciate: 67% of high performers report a “crisis of meaning” when AI can do their job. They’re not afraid of replacement. They’re disgusted by the revelation that their last five years of 60-hour weeks could be replicated by a chatbot in an afternoon.

The talent market is bifurcating faster than you can say “transformational synergies.” On one side: people who want to do work only humans can do. On the other: people content to remain middle managers in the PowerPoint Industrial Complex. Guess which group builds the future.

Which means AI isn’t taking jobs—it’s exposing fake jobs. And your talent is having an existential crisis that no amount of pizza Fridays can fix.

The Contrarian Turn

Here’s what McKinsey won’t tell you but I will: Most AI consultants are selling you aspirin for cancer.

Work slop is a symptom of three deeper diseases:

Incentive Scurvy:

You measure activity, not outcomes. Your KPIs are KP-Lies. You count emails sent, meetings attended, documents created. You’re essentially paying people to move pixels around screens and calling it productivity. No wonder AI excels at this—pointless busy work is computationally trivial.

Process Obesity:

70% of your workflows are organisational cholesterol. They exist because they existed yesterday. They persist because nobody has the courage to kill them. They metastasise because process creation is rewarded but process elimination is ignored. Your company isn’t running—it’s waddling, weighed down by decades of accumulated procedural plaque.

Leadership Anaemia:

Your managers manage tasks, not goals. They’re schedulers, not strategists. They spend 73% of their time in meetings about meetings, creating work about work. They’re not leaders—they’re highly paid administrators with better titles.

And here’s the uncomfortable truth that’ll make your middle management layer squirm: They are the primary slop generation engine. Not because they’re incompetent—they’re often quite intelligent. But you’ve trained them to be “visibility manufacturers” rather than value creators. They produce evidence of effort, not proof of progress.

Here’s my trial or my gauntlet. The test that separates real companies from corporate cargo cults: If your company disappeared tomorrow, would anyone besides employees notice within 30 days? Would customers call asking where you went? Would markets shift? Would competitors celebrate? If the answer is no, you’re not a company—you’re a jobs programme with a marketing department.

For me, AI didn’t create this problem, which means AI can’t solve it. Only you can. And the organisations that figure this out don’t just win—they transcend!

The hard reality?

Firing people won’t fix this—they’ll just create slop elsewhere. Buying better AI won’t fix this—GPT-10 would just create better-looking slop. Only organisational chemotherapy will fix this, and it starts with leadership admitting the patient is sick.

The S.L.O.P. Elimination Protocol™

Here’s your treatment plan. Warning: Side effects include actual productivity, employee meaning, and competitive advantage.

S - Scorecard Revolution

Before: Measuring emails sent, meetings attended, slides created, tickets closed, reports filed.

After: Measuring problems solved, revenue impacted, customers delighted, innovations shipped.

Implementation is brutal but simple. Kill ALL activity metrics within 30 days. Not reduce—kill. Dead. Cremated. Scattered to the winds.

Replace them with outcome metrics. Not “conducted 47 customer calls” but “increased customer retention by 12%.” Not “wrote 15 reports” but “identified and eliminated £2M in operational waste.”

What gets measured gets done. Measure garbage, get garbage. Measure outcomes, get outcomes. This isn’t complicated—it’s just hard because it requires actual leadership.

L - Legacy Process Cremation

Time for the Office Space Test. Would Peter Gibbons do this task? If yes, eliminate it. No committee, no discussion, no gradual phase-out. Delete. Now.

Here’s a method I enjoy: Give AI all your processes—every workflow, every procedure, every recurring task. Whatever it can do entirely without human intervention? That’s unnecessary. You’ve been paying humans to do robot work, and now the robots are here to collect.

Target: Eliminate 40% of processes in 90 days. Sounds impossible? Spotify eliminated 8,000 weekly reports. Not reduced—eliminated. Stock up 23%. Turns out nobody missed the reports because nobody read the reports because the reports reported nothing worth reporting.

O - Outcome Orchestration

Transform every role from “task executor” to “outcome owner.” No more job descriptions that read like grocery lists of activities.

Instead:

• What outcome do you own?
• What number do you move?
• What problem do you solve?

The test: Can the employee explain in one sentence how their work impacts revenue or customers? If no, restructure the role or eliminate it.

Harsh? Yes. Necessary? Absolutely.

New management mantra: “I don’t care how. I care why and what.”

Stop managing people’s time and start managing their impact. The best work might happen at 2am or 2pm—who cares if the outcome is achieved?

My prediction: “Everyone becomes a manager, AI becomes everyone’s team.” Your receptionist becomes a customer experience orchestrator with AI handling the actual calls. Your analyst becomes an insight architect with AI crunching the numbers. Humans do what humans do best—judge, create, connect, decide.

AI does everything else.

P - Performance Theatre Penalty

Institute a “Slop Tax”—time wasted on AI slop gets charged to the creator’s department budget. Watch how quickly people learn prompt engineering when there’s a price tag attached to garbage generation.

Create a game in each team, the “Slop Shame” game—a weekly award for the worst AI-generated waste. Make it visible. Make it memorable. Make it something nobody wants to win. Within professionalism perhaps a silly hat to be warn during Friday lunch. The “Golden Slide Award” for most meaningless presentation. The “Tolstoy Trophy” for longest document saying nothing.

But also reward simplicity. Bonuses for work eliminated, not work created. Promote the person who killed 100 processes, not the person who created 10. Celebrate the manager who reduced their team’s meetings by 80%, not the one who pioneered the double-daily standup.

The Monday Morning Test

Enough theory. Here’s what you do before your 9am coffee.

Immediate Action (No Budget, No Permission Needed):

Cancel your most pointless recurring meeting. Right now. Open your calendar, find the meeting where everyone attends but nobody knows why, and delete it. When asked why, say: “AI could generate these updates. If AI can do it, should humans?” Watch the panic. Document the revelation. That panic is your organisation realising its clothes are made of computational nothing.

30-Day Initiative (With Metrics):

Launch the Work Slop Audit:

• Week 1: Track every document over 5 pages created in your department. For each, ask: “What decision did this change?” If the answer is “none,” you’ve found slop.
• Week 2: AI-generate all regular reports using last month’s data. Compare to human versions. Find the delta. Spoiler: There isn’t one, except the AI’s might be longer.
• Week 3: Calculate cost. Hours spent × hourly rate × futility factor (percentage of work that changed nothing). Prepare for nausea when you see the number.
• Week 4: Present findings to the board with one slide. Not 47. One. If you can’t explain the problem and solution on one slide, you don’t understand it well enough.

Success metric: Identify £500K in eliminable work. If you can’t find it, you’re not looking.

Strategic Question for the Board:

“If our competitors eliminated all work that AI can do autonomously, and redeployed that human capital to work AI cannot do, what would our competitive position be in 18 months?”

Let that question sit. Let it marinate in their consciousness. Because while they’re pondering, your competitors are acting.

The Choice

You have two choices. Blame AI for exposing your organisational dysfunction, or thank AI for the diagnosis and get to work on the cure. Choose wisely—your competitors already have.

The organisations that solve this don’t just become more efficient. They become fundamentally different—more human, more creative, more valuable. They stop cosplaying productivity and start creating value. They stop measuring inputs and start measuring impact. They stop managing tasks and start orchestrating outcomes.

The slop era is ending. Not because AI will get better at hiding the symptoms, but because the organisations that matter will cure the disease. The question isn’t whether your industry will be transformed—it’s whether you’ll be the transformer or the transformed.

Work slop isn’t your AI problem. It’s your organisational confession. The machines didn’t create meaningless work—they just made it impossible to ignore. Now you know. The diagnosis is complete. The prescription is clear. The only question remaining is whether you have the courage to take the medicine.

Your move, Monday morning.

For the S.L.O.P. Elimination Protocol™ diagnostic tool and implementation guide, visit [your-consultancy.com/slop]. For organisations ready to transform, not just talk, reach out. We specialise in controlled demolition of corporate theatre.