Example Oy, 240 employees

Score 0–100 · ▲ change since previous round

1. 1
   Engineering
   72.0▲ 6.0
2. 2
   Finance
   61.0▲ 4.0
3. 3
   Sales
   54.0▲ 3.0
4. 4
   Customer service
   47.0▲ 5.0

• Checking output: Customer service rarely verifies AI answers before use. Measure: short role-specific practice on verification.
• Iterating: Sales accepts first drafts. Measure: prompt-iteration workshop on real sales tasks.
• Data handling: Finance unsure what may be shared with tools. Measure: update AI use policy and brief the team.

One possible high-risk use case identified (AI-assisted CV screening in HR). Two named overseers; one has completed the practical evaluation. Action: evaluate the second overseer and confirm their authority to intervene.

Anonymised peers · grows as data accumulates

1. 1
   Example Oy
   59.0
2. 2
   Organisation A
   66.0
3. 3
   Organisation B
   52.0
4. 4
   Organisation C
   44.0

Each participant completes a 30–45 minute practical task with their usual AI tools. Two dimensions are scored the same way for every organisation: process (prompting, iterating, checking the output) and quality of the result. Scores are aggregated by role group; no individual scores are reported.

Method version 1.0 · Sample generated October 2026