By Marcus, Sales, HUMAN HIVE
Why are senior leaders getting so much more out of AI than their junior teams? Because AI hands back the expertise you bring to it, and juniors have not had the years to build that expertise yet. A new PA Consulting survey of 2,000 workers puts numbers on the gap, and the fix is a people plan, not a software plan.
The 60-second version
PA Consulting found 70% of senior leaders say AI lets them do things they could not do before, compared with 38% of entry-level and junior staff.
The difference is judgment: experienced people know what good looks like, so they can challenge what AI gives them.
This week, make one junior teammate explain why an AI-shaped draft is right before it goes out.
What happened
On November 16, 2026, ITPro reported on PA Consulting's report "The capability paradox: Why human skill and judgement matter more as AI advances". PA built it from a survey of 2,000 workers plus interviews and diary research.
The headline is a split. Among senior leaders, 70% said AI helps them "achieve things they could not have done before," 68% said it lets them "focus on higher-value work," and 72% said they are delivering higher quality work. Senior leaders use AI about 6.9 times a day on average. Among entry-level and junior employees, only 38% reported the same gains in getting work done, and 43% reported higher quality.
Two more numbers should make every manager sit up. Only 12% of respondents said they trust AI's judgement over their own, and 26% said they "feel uncomfortable signing off work that has been shaped by AI." In Raconteur's report on the same PA research, 37% of employees said they spend as long reviewing AI output as they would doing the work by hand, and 34% said they feel less responsible for quality when AI is involved.
Why it happened: the human side
AI did not pick favorites. It reflects the person using it. Dean Collins, people and technology expert at PA, told ITPro that AI "presents back to the individual their own capability and expertise." Someone "who knows what good looks like" can interrogate the answer and push it until it holds up. Someone early in their career often cannot tell a strong answer from a confident one.
Here is the part that should worry leaders most. Collins told Raconteur that "junior employees are missing out on opportunities to build expertise, judgement and confidence through observation, trial, feedback and repetition." The first drafts, the rough research and the meeting notes used to be how juniors learned the job. AI now does a lot of that work. If nobody redesigns the apprenticeship, the skill pipeline quietly dries up.
Other research points the same way. A Microsoft Research and Carnegie Mellon study of 319 knowledge workers, presented at CHI 2025, found that more confidence in generative AI went with less critical thinking, while more confidence in one's own skills went with more of it.
What Mitch has taught on this
Mitch Mitchem saw this coming. On the Event Business Formula podcast in June 2024, he described what he tells young people who are afraid of AI: "you're at a time in history where you're going to be able to operate with the exact or greater level of knowledge than anyone ahead of you." Then came the catch: "what someone won't have is wisdom, which is applied knowledge. They won't have that application. That takes time."
That is the capability paradox in one breath. Knowledge is now on tap for everyone. Applied knowledge still has to be earned. And Mitch's view of the tool is the same one HIVE teaches: "I look at it like AI is an augmentation of us. It uplifts us. It makes us better, smarter." It amplifies what is already there, which is exactly why leaders have to build what is there. Amplify, Not Just Automate.
Best practices for leaders, L&D and sales managers
Add an "explain it back" step to sign-off. Before a junior sends AI-shaped work, they answer three questions out loud: what did AI get right, what did you change, and how do you know it is true? Managers run this in one-on-ones.
Review out loud, not in silence. When a senior person fixes an AI draft, they narrate why for five minutes with a junior watching. That puts back the observation that AI removed.
Keep some reps AI-off. Pick one recurring task per junior role (a first call plan, a summary, a proposal outline) and have them do it by hand first, then compare it with the AI version. Sales managers can do this with discovery call prep.
Name the owner on every output. PA's survey found 34% feel less responsible for quality when AI is involved. Write it into your team norms: the person who signs it owns it, no matter who or what drafted it.
Measure judgment, not usage. Track how often juniors catch an AI error or improve a draft, not how many prompts they run. Praise the catch in public.
What's driving people crazy this week
The cleanup is landing on the experienced people. The Cool Down reported on November 13 that a social services worker said colleagues' unedited AI documents, full of repetition and unintended promises, had "added hours to my workday" and crowded out senior work. That is the capability paradox from the other side. The human fix: make the person who prompts the draft responsible for checking it, and teach them how, so the review does not fall on whoever knows the most.
What this means for the next 1 to 3 years
Apprenticeship gets designed on purpose. Companies that win will treat judgment like a curriculum: paired reviews, shadow sessions and "show your thinking" sign-offs, built into the week instead of left to chance.
Juniors become the best AI testers. Give new hires the job of stress-testing AI answers against real customers and real outcomes. They build judgment fast, and the team gets a check on the tool.
Managers become coaches of judgment. As AI handles more first drafts, the manager's value shifts to asking better questions and teaching others to ask them.
The leadership bench depends on it. Collins warned that today's junior and mid-level staff are tomorrow's leaders. If AI keeps them from failing and learning now, they arrive at the top without the experience leaders lean on.
How we evolve
The human skill to build here is judgment: knowing what good looks like and being able to say why. As AI gets better at producing answers, the scarce skill becomes telling a right answer from a plausible one, and owning the call. That is the heart of Mitch's ADAPT thinking and the Human Element: let the machine bring the knowledge, and let people bring the wisdom.
One small daily practice: once a day, before you accept an AI answer, write one sentence that starts "This is right because..." If you cannot finish the sentence, you have found where your judgment needs to grow.
Try this today
Ask at your next team meeting (5 minutes): "What is one thing AI got wrong for you this month, and how did you catch it?" Let the senior people answer first so juniors see how it is done.
Pick one AI-off rep (10 minutes): choose one task each junior teammate will do by hand this week before comparing it with the AI version.
Add the sign-off check (10 minutes): paste the three questions below into your team's review checklist or project template.
Free tool: the 3-question sign-off check
Before you send anything AI helped create, answer these in writing: 1. What did the AI get right, and what did I change? 2. Which facts, names or numbers did I check against a real source? 3. Would I sign my name to this if nobody knew AI was involved? If any answer is "I'm not sure," ask a teammate before you hit send.
FAQ
What is the AI capability paradox?
It is PA Consulting's name for a gap in how AI pays off at work. In its survey of 2,000 workers, 70% of senior leaders said AI helps them achieve things they could not do before, compared with 38% of entry-level and junior staff. The people with the most experience get the most from the tool, because they know what good looks like.
Why do junior employees get less value from AI?
PA's Dean Collins says experienced people can interrogate AI, test and challenge its answers and judge the result. Newer employees often lack that judgment yet, and AI now does much of the junior work through which they used to build it: observation, trial, feedback and repetition.
How can managers help junior staff use AI well?
Make judgment visible and practiced. Ask juniors to explain why an AI-shaped draft is right before they send it, have seniors review out loud, keep some work AI-off so skills still get built, and be clear that the person who signs off owns the output.
Where HIVE can help
Skills gaps are people gaps. Book a 20-minute call about where your workforce is stuck. Want to build the judgment, listening and presence AI cannot copy? See our human skills training.