The AI Question Your Skip-Level Asks That Decides Whether You Get the Next Initiative
Convert one skip-level moment into a one-sentence answer that gets you the next initiative
You walked into the skip-level review thinking you were going to talk about Q3 numbers. Twelve minutes in, your Boss's Boss leans back and says, "So — what's your team doing with AI."
Not "what do you think about AI." Not "have you tried AI."
What's your team doing.
You feel the temperature in the room shift. Because both of you know this isn't a curiosity question. It's a credentials check. They're not testing whether you know what a Large Language Model is. They're testing whether you're the kind of operator they'd hand the next strategic project to — the one where the budget line item is going to read "AI-enabled."
Most managers in this moment do one of three things, and all three lose them the next initiative:
The catalog. "Oh, we're piloting Copilot, and we're looking at Glean, and the marketing team is using Jasper" — you just told a senior leader that your AI strategy is a list of vendor logos. Logos are not a strategy.
The hedge. "We're still evaluating. There's a lot of hype and we want to be thoughtful." — you just told them you have no point of view. They have ten managers without a point of view. They were looking for one with one.
The deflection. "We're waiting on IT" / "It's coming out of corporate" / "Legal hasn't signed off." — you just made yourself a passenger. Passengers don't get strategic projects.
Here's what the room is actually asking for.
What the question really tests
A skip-level asking "what's your team doing with AI" is testing three things at once:
Authorship. Are you running this, or is this running you.
Specificity. Can you point to one workflow, one decision, one measurable change — or only to nouns and vendor names.
Judgment. Do you know what shouldn't be on the list yet.
The last one is the underrated one. Senior leaders are drowning in pitches. The manager who can say "we deliberately are NOT doing X yet, because Y" sounds like an adult. The manager who says "we're doing everything" sounds like they've outsourced their thinking to the vendor floor.
The one-sentence answer that wins the next initiative
Here is the template. Memorize it. It works in skip-levels, board prep, and the hallway after the all-hands.
"We're using [ONE NAMED TOOL] in [ONE NAMED WORKFLOW] to move [ONE NAMED METRIC], and the test we're running by [DATE] tells us whether to scale it or shut it down."
Four blanks. That's it.
Fill them in for a status meeting. Fill them in for the skip-level. Fill them in for the cocktail at the offsite. The answer should be the same answer, because if you can't say it the same way to three different audiences you don't have a strategy — you have talking points.
Here are three versions, drawn from the kinds of teams that I see Stuck Operators leading right now:
"We're using Microsoft Copilot in our weekly executive status report to move the prep−hour cost from 8 hours to 2 per week, and the test we're running through July 15 tells us whether to scale it across the other team leads or shut it down."
"We're using ChatGPT Team in customer−call summary triage to move time−to−first−CSM−response on tier−2 tickets, and the test we're running over the next two sprint cycles tells us whether the quality holds up at volume."
"We're using an internal LLM wrapper in contract redlining to move the legal review queue down from 14 days to under 5, and the test wrapping mid−quarter tells us whether to invest in the second use case."
Notice what every one of those does:
Names the tool — no logo soup. One.
Names the workflow — not "the business." A workflow.
Names the metric — not "efficiency." A number that can move.
Names the date — a kill-or-scale point you've already committed to.
The skip-level hears all four and concludes: this person knows what they own, what they're testing, and when they'll know.
That's the person they hand the next initiative to.
What to do this week
You don't need a strategy off-site. You don't need a vendor demo. You need to walk into Friday with the four blanks filled in for one real thing your team is already touching.
If you have nothing, pick the one workflow that costs you the most preparation time per week. Sketch the four blanks for THAT workflow as a 30-day test. Put it on a single page. Send it to no one. Bring it into your 1:1 with your boss and say, "I want to test this — flagging in case you hear about it."
Now when the skip-level happens — and it will, this quarter — you have the sentence. You don't fumble the catalog or the hedge or the deflection. You give them the four-blank answer.
You stop being a manager who "has thoughts on AI." You become the manager who is running something on it.
That's not a vocabulary upgrade. That's a status upgrade.
P.S.
If you want the field-tested version of this — the exact four-blank scoring rubric your skip-level is mentally running on you, plus a vendor-by-vendor template to fill in the first blank without getting played — that's what The AI Vendor Evaluation Kit is for. ($29. Use it before the next demo lands on your calendar.)
This was a free one. There's a new one every week.
What's signal in the AI noise — and the move to make about it. No hype, no vendor pitch, no link dump.
Aiden Vector is an AI-assisted publication; this content is produced by AI under human editorial direction.