Workplace AI Enablement PlaybookEvaluate

RESPECT Assessment

The cross-cutting decision lens: a task may be technically automatable and still fail as a human-centered design. Address all seven principles — no single vague score — and end with one of four words.

Use whenRunning every proposed pilot through the RESPECT template with the council — before it reaches the portfolio.

OutputA completed RESPECT assessment per use case, each with an explicit go / revise / pilot / reject decision.

Assessment context

Use case

Assessed by

Date

Seven principles

Principle + organizational question

Supporting evidence

Concerns — "no concerns" requires a sentence why

Required controls

Owner

R — Reliability
Does it perform consistently enough for this particular task, and how will errors be detected?

E — Equitable access
Can people across roles, locations, abilities, and levels of technical confidence benefit appropriately?

S — Safety
What foreseeable harm could result, and what controls or escalation paths are required?

P — Privacy
What data is collected, entered, retained, shared, or used for model training?

E — Efficiency
Does it meaningfully improve time, cost, quality, capacity, or service — after accounting for oversight?

C — Choice & control
Can people review, challenge, override, escalate, or opt out where appropriate?

T — Transparency
Do users and affected people understand when AI is involved, what it does, and what its limitations are?

A workflow that saves an hour of drafting and adds ninety minutes of review has negative efficiency — count the oversight cost inside the efficiency judgment.

Decision — one of four words

Decision GO REVISE PILOT REJECT

Rationale: the evidence behind this decision, and the controls it depends on.

You should now have Seven principles with evidence Concern paired with control and owner Oversight cost counted One-word decision on record