The Cornerstone
Worker

Here is the question almost every leader is being asked right now, usually badly: which jobs does AI replace?

It’s the wrong question. It sends you into a defensive crouch — mapping tasks to models, counting heads, bracing for cuts. You end up managing a shrinking inventory of work instead of building the team that wins the next decade.

The better question is the one a Chief asks before a deployment, not the one a spreadsheet asks before a layoff: who do we build around?

I spent twenty years in the Navy and earned my anchors as a Chief — the rank where you decide who gets the watch, the work, and the trust. I now operate in commercial real estate, where I make the same call with capital instead of sailors. The decision is identical in both worlds. You don’t build around the person who completes the most tasks. You build around the cornerstone — the one whose absence would crack the wall.

AI Takes the Tasks. That’s the Filter.

AI is very good at tasks. It is taking the easy half of nearly every job, and it is taking it fast. McKinsey estimates that today’s AI can automate the activities that fill 60 to 70 percent of an employee’s time, with the heaviest load on knowledge work, not the factory floor. Goldman Sachs ranks office and administrative support as the single most exposed category in the economy — 46 percent of its tasks open to automation — and puts the equivalent of roughly 300 million full-time jobs in the path of the technology. The World Economic Forum’s Future of Jobs Report 2025 projects 92 million roles displaced and 170 million created by 2030, with 41 percent of employers expecting to reduce headcount as AI absorbs tasks, and 39 percent of workers’ core skills changing in the same window.

Read those numbers the way most people do and you brace for impact. Read them the way a Chief reads a roster and you see something else: a filter. When the routine work evaporates, what’s left is exactly the work that made someone a cornerstone in the first place — judgment, initiative, trust earned through execution, the instinct to bring value before being asked. AI doesn’t touch that layer. It exposes who actually had it.

The work that’s leaving was never the work that made you.

So the job of leading people doesn’t get smaller. It gets sharper. Four moves matter now.

01Rewrite the job, not the headcount.

Most job descriptions are task lists. Task lists are now shopping lists for automation. Rewrite roles around outcomes and judgment: what decision does this person own, what gets worse when they’re gone, what would a model still get wrong? If you can’t answer that, you haven’t defined a job — you’ve defined a queue.

02Hire for evidence of value created, not boxes checked.

A résumé is a list of tasks performed. A cornerstone shows up differently. They’ve solved a problem nobody assigned them. They’ve earned access they weren’t given. Build your interviews to surface that — ask for the time they brought something to a room before anyone asked for it. The candidate who has only ever completed assignments is the candidate AI competes with most directly.

03Develop the human layer on purpose.

Organizations spend training budgets teaching people skills the model already has. Flip it. Develop the things that compound and don’t automate: how to earn trust, how to communicate with clarity, how to manufacture an opportunity instead of waiting for one. In The Value Doctrine I make that last one a principle — opportunity is manufactured — because it’s a habit you can build, not a temperament you’re born with. It is also the habit AI is least able to imitate.

04Retain the cornerstones like they’re cornerstones.

When the routine work disappears, your best people get more valuable and more visible to everyone else. Retention stops being a perks problem and becomes an investment decision: who are the dozen people this organization is genuinely built on, and what are you doing to keep them? If you don’t know the names, that’s the first project.

The Most Operational Thing You’ll Do This Decade

None of this is soft. The companies that come out ahead won’t be the ones that cut fastest. They’ll be the ones that identified their cornerstones early, built around them deliberately, and used the time AI gave back to grow more of them.

And if you’re the worker reading this rather than the one doing the hiring, the instruction is the same, pointed inward: stop defending the tasks. They were never the work that mattered. Become the person whose absence would crack the wall. That is a decision — available to you right now, with what you have, from where you are.

Stop defending the work that’s leaving. Start building the thing AI can’t.