Use AI to Build a Manager Readiness Brief: Spot Which Teams Are Most Exposed to AI-Driven Change

The task

You're the HRBP or People Ops lead asked by your CHRO to figure out which people managers are most exposed to AI-driven workload change — and which of them can actually coach their team through it. This usually lands on your desk before a quarterly talent review or a policy rollout. The output is a short brief the CHRO can act on: who needs a coaching plan, who needs a policy nudge, who's fine.

Before AI

Today this is a spreadsheet job. You pull team rosters, cross-reference role descriptions with what you know is being automated, ping skip-levels for a gut check on each manager's AI fluency, and then write it up. Half a day minimum, and it's mostly your subjective read.

The pressure is real: Fifty per cent of the CHROs surveyed said they are not very confident, or not at all confident, in their managers' ability to guide employees on using AI at work. Meanwhile 99% consider AI somewhat or very important to their organization's strategy. If you want context on the split, HR Dive's write-up of the Gallup findings is the cleanest summary.

The workflow

The three prompts below run in sequence. Paste the sample input at the end of prompt 1. Prompts 2 and 3 keep working on the output of the step before.

1. Score each team's exposure to AI-driven change

You are an HR analyst helping a People Ops lead build a Manager Readiness Brief.

Below is a roster of managers and their teams. For each team, produce an "AI Exposure Score" from 1 (low) to 5 (high) based on:
- Share of the team's work that is routine text, data, or code generation (higher = more exposed)
- Share that is judgment, physical, or high-stakes-regulated work (higher = less exposed)
- Whether the team already uses AI tools day-to-day

For each manager, output a table with columns: Manager, Team, Headcount, Exposure Score (1-5), One-line rationale.

Rules:
- Do not speculate about individuals beyond the data given. Reason only from the team's stated work.
- If data is missing for a team, mark Exposure Score as "Unknown" and say what's missing.
- Do not recommend layoffs, PIPs, or terminations. This is a coaching-planning exercise, not a workforce reduction.

Roster follows below.

Company: Northwind Retail (fictional), ~1,400 employees. Readiness cycle: Q3 2026.

Managers and teams:

1) Manager: Priya Alvarado — Team: Customer Support Tier 1 — Headcount: 22
Work: Handles inbound email and chat tickets, mostly password resets, order lookups, refund requests. Uses a canned-response tool. No AI copilots deployed yet.
Manager's AI background: Attended one internal "Intro to Copilot" lunch-and-learn. Has not shipped any AI-assisted workflow.
Recent 1:1 signal (from skip-level): Team is nervous about being replaced; Priya avoids the topic.

2) Manager: Devon Kettering — Team: Marketing Content — Headcount: 8
Work: Blog posts, email campaigns, product one-pagers, social copy.
Manager's AI background: Runs a personal newsletter drafted with ChatGPT. Built a shared prompt library for the team six months ago.
Recent 1:1 signal: Team using AI daily; Devon has published internal norms on disclosure and fact-checking.

3) Manager: Sam Ohtake — Team: Warehouse Ops — Shift B — Headcount: 46
Work: Pick, pack, ship, forklift operation, safety audits.
Manager's AI background: None reported. Uses the WMS dashboard.
Recent 1:1 signal: Team unaffected by generative AI so far; some anxiety about warehouse robotics pilots in other regions.

4) Manager: Rita Bloomquist — Team: Payroll & Benefits Ops — Headcount: 6
Work: Semi-monthly payroll runs, benefits enrollment, tier-2 employee questions, compliance filings.
Manager's AI background: Skeptical publicly; concerned about PII in prompts. Has blocked team from using external LLMs.
Recent 1:1 signal: Team quietly using consumer ChatGPT on personal devices for policy lookups.

5) Manager: Jordan Freed — Team: Data Engineering — Headcount: 11
Work: Pipeline maintenance, SQL, dbt models, incident response.
Manager's AI background: Team piloted GitHub Copilot; Jordan wrote the eval criteria.
Recent 1:1 signal: High tool fluency, but Jordan flags concern that juniors are shipping code they can't debug.

6) Manager: Lin Bakhtiari — Team: Recruiting — Headcount: 9
Work: Sourcing, screening calls, scheduling, offer negotiation. High volume of req load.
Manager's AI background: Attended vendor demos for AI sourcing tools. Has not adopted; concerned about bias in screening.
Recent 1:1 signal: Team overloaded; Lin resisting AI adoption citing legal/compliance risk (fictional NY AEDT-style rule).

7) Manager: Chris Delacroix — Team: Field Sales — Central Region — Headcount: 14
Work: Outbound prospecting, discovery calls, proposal writing, CRM hygiene.
Manager's AI background: Uses AI to draft proposals personally. No team-wide norms.
Recent 1:1 signal: Reps using varied AI tools; inconsistent messaging showing up in customer emails.

8) Manager: Ash Norgaard — Team: Legal Ops — Headcount: 4
Work: Contract intake, NDA review, vendor risk questionnaires.
Manager's AI background: Piloting a CLM tool with built-in AI review. Cautious about hallucinations.
Recent 1:1 signal: Team wants faster turnaround; Ash gating usage behind partner review.

2. Score each manager's AI-readiness and flag the mismatches

Now, using the same roster, add a second table: "Manager AI-Readiness".

For each manager, score 1 (low) to 5 (high) on:
- Tool fluency: has the manager actually used AI in their own work?
- Team enablement: has the manager set norms, prompts, or guardrails for the team?
- Change posture: are they engaging with the shift or avoiding it?

Columns: Manager, Tool Fluency (1-5), Team Enablement (1-5), Change Posture (1-5), Overall Readiness (average, rounded to 1 decimal), Evidence (one line, quoting the signal you used).

Then produce a third table: "Exposure vs Readiness Gaps".
- For each manager, compute Gap = Exposure Score − Overall Readiness.
- Sort descending. Highest Gap = most urgent to support.
- Flag any manager where the team is quietly using AI without sanctioned tools as a POLICY RISK (separate column, Yes/No).
- Flag any manager whose approach could create bias or compliance exposure (e.g., AI in screening, PII in prompts) as a BIAS/COMPLIANCE RISK (Yes/No, with one-line reason).

Do not assign blame. The goal is to identify where HR support is needed.

3. Draft the CHRO brief

Write a one-page Manager Readiness Brief for the CHRO. Use this structure:

**Headline** (2 sentences): Overall state of manager readiness across the sample, and the single biggest risk.

**Top 3 managers to support this quarter**: For each, give (a) why they're top priority based on the Gap score, (b) one specific, low-cost HR intervention (e.g., pair with a peer, sponsor into a cohort, co-write team AI norms), (c) what success looks like in 90 days.

**Policy actions**: 2-3 concrete policy or comms moves aimed at the whole population — driven by the POLICY RISK and BIAS/COMPLIANCE RISK flags. Name the flag that triggered each action.

**What we are NOT recommending**: One short paragraph explicitly stating that this brief does not support performance actions, terminations, or individual bias determinations. It is a coaching and enablement plan.

Tone: direct, no consultant-speak, no bullet inflation. If the data is thin for any recommendation, say so.

Gotchas

• The scores are directional, not diagnostic. A 4.2 vs a 3.8 doesn't mean anything real. Treat this as a sorting tool for your own judgment, not a rating you'd share with the manager.
• Never paste real names or PII. Anonymize before you prompt — use role + team ID. If your CHRO wants named recommendations, add the names by hand at the end from your own notes.
• Bias risk is real when AI touches hiring. Half of CHROs are not confident in their managers' ability to guide employees on using AI at work — which means bias risk is likely underreported in your inputs, not overreported. The recruiting manager flag in the sample is deliberately there; check your own roster for the equivalent.
• The model will overreach. It will sometimes recommend PIPs, reorgs, or terminations if you don't explicitly forbid it. The prompts above forbid it; don't strip that guardrail.
• "Team is quietly using consumer ChatGPT" is a policy signal, not evidence. Verify before the brief goes up.

Time saved

Manual version: ~4 hours to pull data, interview skip-levels, score, and write. AI-assisted version: ~30-45 minutes including your review pass and de-anonymizing at the end. Conservative estimate: 3-4 hours saved per readiness cycle, per HRBP. The bigger win is that you actually run the cycle instead of skipping it — which matters because employees in organizations using AI were less likely to say their workplace culture had remained unchanged over the previous year, so the ground is moving whether you brief the CHRO or not.

What we got when we ran it

We executed every prompt above, in order, against the sample input (model: claude-sonnet-4-6). Outputs below are real and unedited apart from length truncation.

Step 1 output

# Northwind Retail — Manager Readiness Brief ## AI Exposure Scores by Team | Q3 2026 Planning Cycle > Purpose: Coaching and readiness planning only. Scores reflect team-level work characteristics, not individual performance. No workforce-reduction decisions should be drawn from this output. --- ## Exposure Score Summary Table | Manager | Team | Headcount | Exposure Score (1–5) | One-line Rationale | |---|---|---|---|---| | Priya Alvarado | Customer Support Tier 1 | 22 | 5 | Inbound email/chat handling, order lookups, and refund routing are highly routine text tasks—among the highest-exposure workflows for AI copilots—and no AI has been deployed yet, creating significant near-term change pressure. | | Devon Kettering | Marketing Content | 8 | 5 | Blog posts, email campaigns, and social copy are core generative-AI use cases, and the team is already using AI daily with established norms, meaning disruption to current work patterns is actively underway rather than pending. | | Sam Ohtake | Warehouse Ops – Shift B | 46 | 1 | Pick, pack, ship, forklift operation, and safety audits are physical, procedural, and safety-regulated tasks with minimal text/code-generation components; near-term generative AI exposure is low, though automation and robotics are a longer-horizon consideration. | | Rita Bloomquist | Payroll & Benefits Ops | 6 | 4 | Semi-monthly payroll runs, benefits enrollment, and compliance filings involve high volumes of structured, repeatable data and document work, but PII sensitivity and regulatory compliance create meaningful guardrails that temper raw exposure; shadow usage on personal devices

…output truncated at 250 words.

Step 2 output

# Northwind Retail — Manager Readiness Brief (Extended) ## Q3 2026 Planning Cycle --- ## Table 2 — Manager AI-Readiness Scores > Scores reflect observed managerial behaviors and signals only. Sub-scores are independent; a manager can be personally fluent but a poor team enabler, or vice versa. Sub-score definitions: - Tool Fluency: Has the manager demonstrably used AI in their own work? (1 = none reported; 5 = active, sophisticated personal use) - Team Enablement: Has the manager created norms, guardrails, prompt libraries, or structured guidance for the team? (1 = none; 5 = documented, actively maintained) - Change Posture: Is the manager engaging constructively with the AI transition, or avoiding/blocking? (1 = active avoidance/blocking; 5 = proactive, transparent leadership of the shift) | Manager | Tool Fluency (1–5) | Team Enablement (1–5) | Change Posture (1–5) | Overall Readiness (avg) | Evidence | |---|---|---|---|---|---| | Priya Alvarado | 2 | 1 | 1 | 1.3 | Attended one lunch-and-learn, has not shipped any AI workflow, and skip-level confirms she actively avoids the topic with her team. | | Devon Kettering | 5 | 5 | 5 | 5.0 | Personally drafts newsletter with ChatGPT, built a shared prompt library, and has published internal disclosure and fact-checking norms. | | Sam Ohtake | 1 | 1 | 3 | 1.7 | No AI background reported; no team norms exist; posture is neutral rather than avoidant—team is unaffected and Sam shows no resistance. | | Rita Bloomquist | 1 | 1

…output truncated at 250 words.

Step 3 output

# Manager AI Readiness Brief Prepared for: Chief Human Resources Officer Cycle: Q3 2026 | Population: 8 managers, ~120 employees --- ## Headline Across the eight managers assessed, readiness is uneven in ways that track closely to seniority and personal AI experimentation—managers who have used these tools themselves are leading their teams through the shift; those who haven't are either avoiding the conversation or blocking adoption in ways that are creating new risks rather than containing them. The single biggest risk is not slow adoption: it is that employees in at least two teams are already using consumer AI tools on personal devices with no guardrails, meaning PII and brand exposure is happening now, below the line of manager visibility. --- ## Top 3 Managers to Support This Quarter 1. Priya Alvarado — Customer Support Tier 1 | Gap: +3.7 Priya leads the team with the highest AI exposure in the sample and the lowest readiness score. Her team handles high-volume, routine text work—exactly the workflows AI copilots will touch first—and the skip-level signal confirms the team is already anxious and looking to her for direction she isn't yet providing. This is not a performance issue; it is a preparation gap, and the window to get ahead of it is closing. The intervention is simple and low-cost: pair Priya with Devon Kettering for two structured peer conversations this quarter. Devon has already built what Priya needs—a team communication approach, disclosure norms, and a working prompt framework—and peer transfer is faster

…output truncated at 250 words.