Use It or Lose It: Deciding Which Skills Must Survive AI Delegation

Edmundo Barrientos · 2026-10-21 · 3 min read

Use It or Lose It: Deciding Which Skills Must Survive AI Delegation

TL;DR: When you hand tasks to AI, practice stops and key skills quietly fade. Name the skills people must keep. Protect them with deliberate practice, not total delegation.

The feedback note Maya couldn't draft on her own

Maya became a team lead in March. Tough feedback felt daunting, so she asked AI to write all her notes. She tweaked each draft before sending it. Her team responded well.

That changed in October, in a live team meeting. A colleague missed a deadline for the third time. There was no chance to prompt a tool. Maya had to speak on the spot.

She faltered. Then she said, "Let's discuss this later."

Skill loss is natural when practice stops

Maya's trouble was normal. Skills fade when you stop using them.

Writing in Chief Learning Officer, learning strategist Treca Bourne cites The Week: deskilling isn't inherently negative. She asks how much is appropriate, in what contexts, and what safeguards apply.

At Maya's company, nobody asked. Each delegated draft saved time. No one chose to let her feedback skill fade. It faded because no one decided.

Protecting essential skills: a three-question test

Not every skill is worth saving. Ask three questions about any task you plan to delegate:

  1. Is it critical when AI fails or makes mistakes? Maya must spot a draft that is polite but never names the problem.
  2. Is it needed when AI is not there? In-person talks, live meetings, and phone calls come with no draft.
  3. Does future growth depend on it? A lead who gives good feedback today can coach other leads tomorrow.

One "yes" marks a core skill worth keeping. Bourne makes the first two points with surgeons who still operate without a robot, and pilots who practice manual landings. In her words, "Automation can extend human capability, but resilience depends on preserving essential human expertise."

Keep the list short. A list of dozens protects none.

Skill retention requires continuous practice, not reliance on delegation

Keeping a skill does not mean avoiding AI. Maya still uses AI for everyday notes.

Every week, however, she does a short role-play: a conversation with an AI that plays her colleague. It pushes back and goes quiet when she hesitates. She says what she would say, then tries again.

The practice changes with each person. It targets the moments where Maya softens her message. The AI plays the part. Maya is in control: she edits and decides what she says. This is augmentation, not automation.

The true test happens in real time

The weekly drill is not the test. The test came when a colleague again missed a deadline in front of the group. Maya named the problem in two clear sentences.

Skills practiced and observed this way show up in what people do. Across a team, they add up to organizational capabilities. That is capability-building: let AI draft, keep the core skill, and check it survives.

References


About the publisher: this article is published by Grasply (grasply.ai), a capability builder. Grasply turns what a company already knows, its docs, calls and recordings, into skills people use at work, for employees, partners and customers. Practice turns knowing into doing, and Grasply shows proof that the work changed.