Every responsible conversation about AI ends with the same advice: check the output. But being 'the human in the loop' is often hard, draining work, and it gets harder when everything AI produces looks like an end product.

In this week's episode of The Mindtools L&D Podcast, guest host Julie Dirksen and Claire are joined by Dr Markus Bernhardt of Endeavor Intelligence to explore the human behaviors behind validating AI outputs, and what organizations need to do to support them.

We discuss:

• why being the human in the loop requires psychological stamina
• how the agentic 'hamster wheel' erode our ability to validate outputs
• triaging tasks by how close they sit to your core expertise and values

Julie's new book on skills development is Design for How People Build Skills.

For more from Markus, visit endeavorintel.com. For more from Julie, visit usablelearning.com.

In 'What I Learned this Week', Claire shared a new term from a scuba-diving friend: the 'incident pit'. It describes how one small problem, like a cracked mask, escalates into a bigger one.

Julie recommended The Score: How to Stop Playing Somebody Else's Game by C. Thi Nguyen. It looks at how scoring systems, from climbing grades to university rankings, can quietly change what we value.

For more from Mindtools Kineo, visit mindtools-kineo.com. There, you'll find out how we can help your organization build AI skills, and how our Skills Practice scenarios help build human skills.

Connect with our speakers

If you'd like to share your thoughts on this episode, connect with us on LinkedIn:

• Julie Dirksen
• Claire Gibson
• Dr Markus Bernhardt

AI can make a rough idea look finished in seconds. That's great for productivity, but it's a real problem for anyone whose job is to judge the work. If polish, speed and thoroughness can all be generated, how do we know who actually knows their stuff?

In this week's episode of The Mindtools L&D Podcast, guest host Julie Dirksen and Cammy Bean are joined by Koreen Pagano, CEO of Talent Rewire and co-author of Rewiring Your Organization for AI, and Dr Alicia Sanchez, Chief AI Officer at MPF Federal, to explore how we evaluate people's work when AI is part of how it gets done.

We discuss:

• why the 'judgment layer' for AI output is missing in most organizations
• how 'polish' and speed change the way we review work
• what managers should, and shouldn't, evaluate when employees use AI.

Julie referenced a study from Atlassian's Teamwork Lab, which found that AI polish makes it harder to spot problems in a draft, and that a simple 'Early Draft' label largely fixes it.

Cammy mentioned an episode of The Curiosity Shop, in which Brené Brown and Adam Grant discuss building integrity in the age of normalized cheating, including when using AI counts as cheating.

Alicia discussed a lesson from Koreen's book, Rewiring Your Organization for AI, co-written with Kristi Broom: decisions about what to automate belong with middle managers, not only with frontline employees.

Julie's new book, Design for How People Build Skills, explores why skill-building comes down to feedback, and to helping people own their own development.

For the latest from Koreen and her team, visit Talent Rewire.

In 'What I Learned this Week', Julie shared how steel drums are made: old oil drums, pounded into shape and tuned. She also shared her first go at using AI as a travel planner for a trip to Rome. The verdict so far? She's still verifying most of it. For tips, Cammy recommended David Kelly's post on what a personal trip to London taught him about AI and learning.

For more from Mindtools Kineo, visit our new website mindtools-kineo.com. There, you'll find out how we can help your organization build AI skills, and how our Skills Practice scenarios help build human skills.

Connect with our speakers

If you'd like to share your thoughts on this episode, connect with us on LinkedIn:

• Julie Dirksen
• Cammy Bean
• Koreen Pagano
• Alicia Sanchez

At Learning Technologies this year, Nigel Paine said AI is turning us into idiots. It was a glib line, and he admits it. But if we hand our thinking over to a chatbot and accept whatever comes back, is he wrong?

In this week's episode of The Mindtools L&D Podcast, Ross G is joined by Dr Nigel Paine, researcher, author, podcaster, broadcaster and consultant (he's a busy guy), to explore when AI makes us sharper, when it makes us duller, and what organizations should do about it.

We discuss:

• Why 'cognitive offloading' becomes a problem when we offload the thinking that matters
• Arguing with AI, and asking it to 'destroy' your conclusions rather than confirm them
• Why organizations need to rethink entry-level roles and recruit for critical thinking

Nigel draws on Daniel Kahneman's Thinking, Fast and Slow to explain why our brains favor the easy answer.

To find out more about Nigel's work, visit nigelpaine.com or email hello@nigelpaine.com.

In 'What I Learned this Week', Nigel recommended Kevin Fong's new book Minute Six, about what teams do when their plan meets reality.

He also shared research suggesting the aging brain isn't declining, it's adapting, trading memory for specific details for big-picture knowledge.

Ross G recommended the Charles Dickens Museum in London, and its audio walking tour of the streets Dickens knew.

Rolling out enterprise AI licenses? Don't miss out on the benefits they promise. Find out how our AI Skills Academy builds AI skills, including critical thinking, at mindtools-kineo.com, or email Ross G at custom@mindtools.com.

Connect with our speakers

If you'd like to share your thoughts on this episode, connect with us on LinkedIn:

• Ross Garner
• Nigel Paine