# Responsible AI in Recruitment: practice workbook
Talent Engineering Academy | An education initiative by Vitae
Edition 2026.09 | Updated 2026-09-24
Course: https://talentengineering.org/courses/responsible-ai

Define accountable AI use, map risks and make oversight, transparency and challenge routes operational.

## Your deliverable
An AI use-case register and risk-control plan.

Use fictional information. Keep your completed work in your own secure notes. Exercises and capstone work are self-directed, not independently assessed.

## 1. Purpose, users and affected people

Your notes:



## 2. Decision influence and prohibited uses

Your notes:



## 3. Risks, severity and affected groups

Your notes:



## 4. Controls, owner and evidence

Your notes:



## 5. Human authority and challenge route

Your notes:



## 6. Approval scope and review triggers

Your notes:



## Lesson exercises

### 1. Inventory use cases by real influence

Create three use-case records with inputs, output influence, owner and prohibited uses.

Your response:


Worked example: A tool sold as an assistant is routinely used to rank applicants without independent review. The register describes the actual ranking use, because that is what affects candidates.

Self-review guidance: One record should show a low-consequence drafting task and another a consequential decision support task. Describe actual operation rather than repeating marketing terminology.

### 2. Map risks to concrete failure scenarios

Write five risks for one fictional use case, each with a control and verification method.

Your response:


Worked example: A summariser omits career context and makes a candidate appear less experienced. Controls include source-linked claims, reviewer access to originals and sampling for omission, not just a spelling check.

Self-review guidance: Each control should have an owner and observable evidence. Include one risk caused by human overreliance rather than model error alone. State unresolved risks instead of declaring the system risk-free.

### 3. Make oversight and transparency meaningful

Design a reviewer checklist and a candidate-facing explanation for a fictional assistive tool.

Your response:


Worked example: A recruiter receives a ranked list with no underlying evidence and thirty seconds to approve it. The redesign provides criterion evidence, removes automatic rejection and gives the reviewer a genuine escalation route.

Self-review guidance: The reviewer must be able to stop or change the action. The explanation should identify purpose and contact route. Avoid the claim that a human click eliminates all risk.

### 4. Approve a bounded use and revisit it

Write an approval record with scope, five conditions and three re-review triggers.

Your response:


Worked example: A supplier adds an emotion-scoring feature to an approved transcription product. The old approval does not automatically cover the new feature. The owner disables it pending separate assessment.

Self-review guidance: Include model or feature changes, new data types and serious incidents. State who can suspend the use. A procurement signature is not permanent approval for every future capability.

## Portfolio review
Check that your work is internally consistent, distinguishes facts from assumptions, names decision owners and explains its limitations. Revise gaps before using the method in real work.

## Further reading
- [GOV.UK: Responsible AI in recruitment](https://www.gov.uk/government/publications/responsible-ai-in-recruitment-guide/responsible-ai-in-recruitment): UK guidance on procuring and deploying recruitment AI.
- [NIST: AI Risk Management Framework](https://www.nist.gov/itl/ai-risk-management-framework): Voluntary framework for organising AI risks and controls.
- [European Commission: AI regulatory framework](https://digital-strategy.ec.europa.eu/en/policies/regulatory-framework-ai): Check the current rules, scope and implementation dates for EU operations.

Original educational scenarios. References provide further reading and do not imply endorsement. Check current official rules and appropriate professional advice for real legal, financial or regulated decisions.