Hey HR folks!👋

Remember when AI was going to give everyone their Fridays back? Yeah. About that. 😬 

New data shows employees are spending real hours each week fixing low effort AI output that looked polished but was not ready for prime time. The tool that was meant to save time is quietly creating rework.

For HR and people leaders, this is a productivity story, and the math deserves a closer look. 👀 

Today’s edition is presented by SelectSoftware Reviews

Get personalized guidance from an HR software expert and a shortlist of tools that fit your team’s needs. The SSR HR Software Advisor helps you move forward with clarity and confidence, without unnecessary steps or back-and-forth.

👇️ Coming up

In today’s edition

🔁 The productivity math behind AI doesn’t work

🛋️ The break room: Is AI actually saving your teams time?

📚 Human readsources: Cybersecurity scare, AI hiring compliance risks, and a high-profile pregnancy discrimination lawsuit put employers back under the legal microscope.

💭 Opening thoughts

🔁 The productivity math behind AI doesn’t work

🤯 66% of employees say they spend up to 6+ hours a week fixing AI related errors.

That is almost a full workday. Gone. To cleanup. 🧹 

At the same time, organizations are pushing AI adoption hard, expecting leaner teams and faster output. Instead, many employees are becoming part time editors of machine generated drafts.

So where exactly are those “time savings” going? Executives are under pressure to prove AI is paying off. In many companies, employees are encouraged to use AI broadly, often without detailed guidance on what good output actually looks like.

More drafts are flying around. More content. More code. More decks. Also more sighing. 🫠 

For easy reading

🧠 Let’s unpick

What the data shows

Low quality AI output is not rare. Only 39% of employees say it is completely unacceptable and corrected. Everyone else works in environments where it is tolerated, overlooked, or even accepted if deadlines are met.

And when it lands in someone’s inbox, it usually does not get sent back. 49% say they fix the issues themselves instead of escalating or rejecting the work. Two thirds are spending up to six or more hours every week correcting AI mistakes ⏳

The impact is not subtle. 70% say this dynamic increases stress. 67% say it hurts productivity. 65% say morale drops. More than half link it to burnout risk.

In a survey of 1,150 full time workers, 41% could recall receiving a specific instance of low effort AI work that affected them. More than half admitted they have sent low quality AI output at least some of the time. One in ten said that half or more of what they sent was unhelpful.

This is a loop. People receive rushed AI output. They feel stretched. They rush their own AI output. The cycle continues 🔁

What HR and people leaders should pay attention to

When output expands but review capacity does not, someone absorbs the cost.

If employees feel pressure to “use AI” without clarity on what finished, high quality work looks like, usage becomes a performance signal instead of a performance enhancer. Drafts get generated quickly. The thinking and refining happen later.

If rework time is invisible, leaders will overestimate productivity gains and underestimate workload strain. Workforce plans tighten. Expectations rise. Engagement data dips. Teams feel busy and behind at the same time.

The issue is not that AI makes mistakes. The issue is that constant cleanup is starting to feel like part of the job.

You can read more at...

🔗 AI “Workslop” Is Costing Workers Hours Each Week, New Report Finds

🔗 Why People Create AI “Workslop”—and How to Stop It

🎬 Lights, camera, action!

Takeaway (and try this 👇)

If you’re in early discovery mode, here are practical ways to start exploring the 4 day work week operating model:

1. ⏱️ Start asking one simple question in workforce reviews: how many hours are we spending fixing AI output?

2. 📊 Compare AI usage data with productivity data. More drafts does not automatically mean more value.

3. 🧠 Invest in better prompting and review skills so employees are not stuck rewriting everything from scratch

4. 🔍 Look for hidden rework in engagement surveys. Stress and frustration spikes often trace back to workflow friction

5. 😅 Be honest about ROI. If teams are spending six hours a week cleaning up, that belongs in the productivity calculation

👀 Too long didn’t read

TLDR

AI is cranking out more output, but many employees are losing hours fixing it. The time savings story falls apart once you count the editing.

☕The break room

What’s your take?

📚 Additional reading

Human Readsources

1. IDMERIT denies fake data breach claims amid extortion attempt – A global ID verification firm says rumors of a “data of billions” leak were part of a ransomware style extortion effort, not an actual breach.

2. Amazon hiring assessment may qualify as illegal lie detector test – A Massachusetts judge ruled that Amazon must face claims its “workstyle” assessment could meet the legal definition of a lie detector under state law.

3. Google sued over alleged pregnancy discrimination and FMLA retaliation – A former engineer claims she was downgraded and terminated after pregnancy related leave and a request for protected medical leave.

That’s it for today.

Thanks for reading to the end and we hope today’s edition sparked some new ideas for your workplace! 🧠

We know you’re super busy and really appreciate you saving some room for us in your inbox 😀