A few weeks ago I read a stat that stopped me: 86% of finance executives now say AI skills training is worth more than an MBA. That's not a hot take from some AI hype account. That's a PwC survey, covered by CFO Brew, of the people who actually run finance departments.
I'll be honest with you about what that means. The mechanical stuff an MBA teaches — financial modeling, building pitch books, living inside a spreadsheet — is genuinely at risk. AI does a lot of that faster than I do, and I've been doing it for twenty years.
So here's the real question, and it's the one this whole issue is built around: was that ever what an MBA was actually for?
The numbers behind the headline. Since you're a subscriber and not just scrolling past a post, here's the part that didn't fit on LinkedIn: in that same PwC survey, 91% of finance leaders said they're raising compensation for AI skills, and 58% said they're paying a real premium for people who have them. That's not theoretical. That's payroll, already moving.
And it's not just finance. Google's AI & Economy ATLAS report — released this July, built from 15 million real interactions with Google's AI products — found that AI has touched 68% of occupations in some way. But here's the part that matters: it's only being used for about 21% of the actual tasks inside those jobs. Right now, for most people, AI is still an assistant. Not a replacement.
Nevada tells the same story from a different angle. A recent DoubleTrack report found only about 26% of Nevada businesses are using AI at all, and the state has just 7.1 AI specialists for every 1,000 jobs.
That's not a “the jobs are disappearing” story. That's a skills-gap story — and it lines up exactly with what PwC found in finance. The premium isn't going to people with a credential. It's going to people who know how to actually use the tools.
Andy Boettcher, DoubleTrack's chief innovation officer, put it about as plainly as it can be put: AI is not coming for your job. The people who've learned to use it well are the ones coming for jobs — in every industry, not just finance. Which is really the same finding as the PwC survey, just said out loud.
Here's the part I want to say directly to you, because every other AI newsletter is written for 25-year-olds who grew up with this stuff and I'm not writing that newsletter.
If you're 50+ and you've been assuming this train already left without you — it didn't. Boettcher's whole point cuts the other way: you don't need to go back to school for this, and you don't need to have grown up with it. The gap he's describing isn't a credential gap. It's a “has anyone actually tried” gap. That's a gap experience closes faster than a fresh degree does.
The Reframe
Here's a habit I picked up that costs nothing and takes two minutes: before a big decision, I ask AI to argue the strongest possible case against me. Not a gentle devil's advocate — the real case, the one a sharp critic would make if they wanted me to lose.
It's the Harvard case method, minus the classroom and the tuition. That's what “strategic thinking, amplified” actually looks like in practice. Not AI doing the thinking for you — AI giving you a sparring partner most of us never had access to.
Here's a smaller, less dramatic version of the same move, for anyone whose stakes aren't boardroom-level this week.
I'd been automating a manual process with Alteryx and kept hitting the kind of errors that are murder to catch by eye — an extra space in a column header, a field with inconsistent spacing.
Instead of eyeballing rows for an hour, I handed a month of manual output and the Alteryx output to Copilot and asked for a full accuracy breakdown. It came back with a 95% match rate, flagged every single mismatched record, and had the root cause identified in minutes.
Same underlying move as the devil's-advocate prompt. AI as thought partner, not task-doer. It just showed up in a spreadsheet instead of a boardroom. Most of us have a recurring task exactly like this one sitting somewhere, half-cleaned-up, that we've never had time to fix.
The Exclusive: The Board Prep Framework
This part hasn't been on LinkedIn. It's newsletter-only.
I came across this story in Geoff Woods' book The AI-Driven Leader, and it's stuck with me. Woods advises a company with an activist board — the kind of board where quarterly meetings are genuinely brutal, and the team is working toward an exit in the next two years. The board had become a real distraction from actually running the business.
So Woods had the executive team try something:
1. Profile each board member. Have AI interview the team about each board member individually, and write a personality profile for each one.
2. Simulate the meeting before it happens. Feed those profiles plus the actual quarterly deck to the model, and have it play out how each board member is likely to react — surfacing the weak points before the real meeting does.
3. Fix the deck. Iterate on the presentation based on what the simulation exposed.
4. Close the loop. After the real meeting, transcribe it, feed the transcript back to the AI, and have it compare what it predicted to what actually happened — sharpening each board member's profile for next quarter.
The result isn't just a better meeting. It's a system that gets sharper every single quarter instead of starting from zero each time.
And this isn't only useful if you sit on a board. The same framework works for:
• A tough conversation with a boss or a client
• A job interview at 50+, walking into a room that skews thirty years younger than you
• A family conversation about money, retirement, or technology that you've been putting off
You don't need the full four-step version to start. Here's the simplified version I'd actually try this week: before your next high-stakes conversation, ask AI to role-play the other person using what you already know about their priorities and how they communicate — then ask it what you're not seeing.
Why I'm recommending this book
The thesis of The AI-Driven Leader is really the same point I made above about the devil's-advocate prompt, just built out into a full framework. Woods' own summary of it: “You are the Thought Leader, AI is the Thought Partner.” You keep the judgment. AI gives you a sparring partner most executives never get — because everyone around them is usually too busy agreeing.
His broader argument is that AI earns a place next to your leadership, not in place of it — and that no matter how good it gets, it doesn't replace the human parts of the job: the relationships, the trust, the read on a room.
For what it's worth, given how skeptical this audience rightly is of business-book hype: Woods was Chief Growth Officer at Jindal Steel & Power, where he helped take the company's market cap from $750 million to over $12 billion in four years. The book is a #1 international bestseller, with endorsements from executives at Costco, Medtronic, FedEx, and Tesla, and it holds a solid (not suspiciously inflated) 3.91 out of 5 across more than 2,000 Goodreads ratings.
What I like about it: he never treats AI as the smart one in the room. You're still the Thought Leader. AI's job is to be the sparring partner most people never get. That's the version of “AI skills” worth having — not a replacement for the kind of thinking an MBA is supposed to teach. Ammunition for it.
One Thing to Try This Week
It doesn't have to be a board meeting.
Ask AI to help plan this week's dinner menu and build the shopping list from it. Or hand it the spreadsheet you track your portfolio in and ask what it notices. Pick one thing — work or personal, doesn't matter which — and try treating it like a thought partner instead of a search bar.
Hit reply and tell me what you tried. I read every response.
Next week: context — arguably the single biggest factor in whether AI actually helps you or just hands you generic answers. See you Tuesday on LinkedIn.
— Eric
More of this on LinkedIn: linkedin.com/in/eric-skelton-mba-3079a47