Holding Back the Robots? Rethinking How Next-Gen Workers Learn in an AI World

TL;DR: Racing to roll out AI to every desk can short-circuit the learning cycles juniors need to build real judgment. AI strategist François Candelon advises starting with fundamentals, then adding technology. Microsoft’s 2025 Work Trend Index reveals that 82 percent of leaders plan to lean on digital labor in the next 18 months, while youth unemployment sits at 9.4 percent as of March 2025, nearly triple the overall rate. This article unpacks how these stats can cause some serious workforce issues, how over-reliance on AI dulls judgment, and what LinkedIn leaders say about balanced access versus blanket bans.

The recommendation? Make talent before tech your rule. Map the core skills juniors must sweat through, then let them unlock AI in stages with tutor-style GPTs, prompt reviews, and judgment checks. Protect the messy learning moments today so your bench can guide the bots tomorrow.

I had the amazing opportunity to be part of a conversation with François Candelon, who worked as the Global Director at BCG’s Henderson Institute and is one of the most respected voices in AI strategy. This came as one of the many perks of being a member of the Women Love AI Marketing Community.

I asked him something that’s been on my mind: As organizations scale AI, how do we ensure junior employees still get the foundational experience they need to grow? We learn by doing, and if AI takes over the doing, what happens to that learning?

He replied that he recommends intentionally trying to hold back AI tools from junior employees, at least at first. 

François’s response jolted me and instantly made me rethink how I might roll AI out to a team.

Talent Before Tech

Talent first. AI second.

The reasoning? Without hands-on experience and deep domain understanding, those early-career professionals can’t yet evaluate or supervise AI-generated work effectively. AI can amplify talent only after that talent exists.

In a business climate obsessed with “AI for everyone,” this advice flips the script. Companies are racing to upskill everyone at lightning speed; François says pump the brakes.

I felt an instant YESSS in my gut. It makes sense. Every day, I coax ChatGPT for quality, and it still takes a pile of context and a few “try again” moments to get a usable end product. If good writing were a pizza, AI would be the dough mixer. You still need to toss, top, and bake. That guidance comes from years of hits, misses, and face-plants.

Microsoft’s 2025 Work Trend Index Report dives deep into the impact AI is having on organizations, workflows and structures as a whole and analyzes survey data from 31,000 workers in 31 countries, LinkedIn labor market trends and Microsoft 365 productivity data. The report explains,

“To work effectively with agents, all employees will need to adopt a thought partner mindset and build related skills: learning to iterate with AI, knowing when to delegate to AI, prompting with context and intent, refining outputs instead of accepting first drafts, spotting weak reasoning or gaps, and knowing when to push back or steer the conversation or plan.”

In the context of junior employees, none of that is possible without baseline expertise. How would a fresh new hire know when an answer is off, when to refine, or when to scrap and restart?

We’re already swimming in a sea of lackluster content and work, resulting from professionals using AI as a crutch to spare them from thinking rather than a counterpart to push them and help them become better.

François’ stance is a lot to unpack, so I asked my LinkedIn network for their take. Here's their perspectives, and some fresh data.

Voices From the Thread: A 360° View

LinkedIn weighed in quickly, and the thread revealed numerous great insights. Here’s three top takeaways leaders can’t ignore. (Be sure to check out the full thread for all the gems shared.)

Caution: Judgement Gap Ahead

“We’ve been obsessively talking about 'artificial' intelligence for 2 years, but now, frankly, I tend to call it the intelligence of laziness. If used too young and without moderation, AI numbs our critical thinking, deductive abilities, and analytical skills – the very foundation of the human brain. If that could be manageable for our generation, I truly think it could be a disaster for the next.” – Karine Abbou, Community Founder & AI Content Strategist

I 100% agree. I already see AI used as a crutch far too often. Using AI constructively takes discipline, and let’s face it – the temptation to hit the easy button is strong. At best, over-reliance gives us a grayscale world of “blah.” At worst, it leaves an entire workforce, government, and society unable to make crucial calls.

Imagine if your city’s infrastructure team approves an AI-generated maintenance schedule, but can’t see the flaw that leads to a bridge closure and months of gridlock, or a brand new marketing coordinator launches AI-written copy and triggers a cultural backlash because no one learned to vet the nuance. Unless we build in guardrails that nurture critical thinking, we’ll end up swapping human insight for mindless autopilot.

Vanishing Entry-Level Roles

“One big problem is entry-level jobs are evaporating because AI tools are making the people on teams that have rolled out AI massively more productive. Unemployment is low unless you are entry-level. The unemployment rate for individuals aged 16-24 in the United States was 9.4% in March 2025. Compared to the overall unemployment rate, which was 3.6% in March 2025.” – Ellen Thompson , Founder @ Respage

This tracks with recent reports around “quiet AI layoffs,” described as jobs quietly absorbed by tools rather than humans or never posted at all because an AI model can now do the work. You can read more about this in Mike Kaput 's blog AI Layoffs Are Already Here. But Don’t Expect Companies to Always Admit It.

This is a real problem, and will hit some industries and jobs faster than others. But eventually, it will touch and alter every industry and role in some way. We would be naive to say it wouldn’t.

I have to be honest, I mean, come on… this is Life Unfiltered. I’ve already done more with less using AI for something I would have paid for in the past. I used ChatGPT to help me come up with a style guide and color scheme for my blog, and I’ve started using it to create on-brand images for my personal brand. Passion project or not, that’s income someone didn’t earn..

Microsoft’s report echoes this shift, citing

“82% of leaders say they’re confident that they’ll use digital labor to expand their workforce capacity in the next 12-18 months.”

The report attributes this to a capacity gap where “53% of leaders say productivity must increase, but 80% of the global workforce – both employees and leaders- say they’re lacking enough time or energy to do their work.”

On a balance sheet, fewer hires look great today. But in three years, when those missing juniors should be stepping into mid-level seats that AI can’t fill, who’s the successor? No bench = no succession plan for your company + a lot of risk.

Balanced Access, Not Blanket Bans

Right as I was ready to go all in on no AI tools for junior employees, • Aleksandra Pimenides popped in with a voice of reason.

“Having junior employees develop core skills and be self-sufficient is crucial, but a blanket AI tool restriction for them isn’t ideal; it could waste time on tasks that could be streamlined or even prevent learning. It depends on the AI tool/its use.

Rather than risking throwing the baby out with the bathwater, let’s ensure juniors use AI wisely, focusing on what they truly must learn, while leveraging AI to boost efficiency/learning.”

– Aleksandra Pimenides, Marketing Strategist

Aleksandra shared an example of a junior employee using AI to learn Excel functions relevant to a task. Using AI, the employee was able to learn the specific skill needed in one hour, versus spending five hours stumbling through tons of applications she didn’t need. She makes a solid point.

Our challenge as leaders is to separate busywork from skill-building. We must be VERY intentional in how we approach this, and likely need to look to data and research. I have a nagging sense that there are a lot of things I do and know that I’ve picked up subconsciously along the way, like the trick you discover while wrestling a spreadsheet or the feature you stumble on while fixing a mistake. Lose those struggle moments, and we may lose the tacit know-how that surfaces later when high-stakes judgment is required. We might not notice the gap until it shows up in a crisis. This scares me.

I agree with Aleksandra. A blanket ban on AI is the wrong move. Finding the right balance between what to automate, and what to sweat through is quickly becoming one of the hardest leadership puzzles we face.

Practical Takeaways For Leaders 

I haven’t cracked the code on this workforce shift. Writing is how I think it through in real time and surface the questions that keep me up at night. We’re all learning on the fly, so below is my first-draft checklist, shaped by recent research and our LinkedIn discussion. Jump into the comments with your tweaks and additions so we can refine it together.

1. Reflect and pin down the core skills. Think back on your career journey. What are the required core competencies each role must master the hard way? Clean data, customer calls, conducting research for a pitch?
2. Study how we learn. Explore research to understand how we learn. Then identify what should remain manual for newbies and where they can use AI without killing judgment. Buddy up with a friend with learning and development expertise. I’m sure they can help.
3. Gamify AI unlocks. A fantastic recommendation from Devin Jacob. Make it fun! Give junior employees milestones to hit to “unlock” AI and feel the win.
4. Use AI as a tutor. Create role-specific GPTs that quiz, coach, and verify skill checkpoints before handing over the shortcuts.
5. Evaluate and monitor for judgment, not just output speed. Review prompts, push back, question reasoning, and mandate that a human reviews all final outputs.

We’re writing the new playbook for how work happens. We should probably write in pencil because it’s going to change frequently. AI can turbo-charge our performance, but only if we guard the messy, hands-on moments that turn newbies into pros. The more openly we share lessons and work together, the faster we build a workforce that can guide AI instead of leaning on it.

What’s one foundational skill you’ll protect before handing junior employees an AI shortcut?

Originally published at ReneeMcintyre.com