Skip to main content
Hafan

Understanding how meaningful human involvement applies: use cases

Contents

In this brief section, we provide some indicative use cases to help demonstrate how the ADM provisions applies in a recruitment context. They focus on how and why it is likely that ADM applies.

Applying meaningful human involvement

For use of an automated recruitment tool to be considered decision support (and not ADM), employers must build meaningful human involvement into each stage of the process for making recruitment decisions about candidates. 

This means that a human must make every decision about whether to progress a candidate to the next stage. If employers use automation to support decision-making, rather than choosing to use ADM, they should ensure that a human: 

  • assesses and reviews the decision at an appropriate point to ensure they have an actual impact on whether a candidate is selected or eliminated from the recruitment process;
  • has the ability to influence the recruitment decision;
  • has the discretion and and authority to alter the recruitment decision;
  • has the qualifications and training to disagree with the tool’s recommendations or predictions and can overturn them;
  • where multiple candidates have similar qualifications and experience, a human makes the decision about who to interview (although they can consider the software’s recommendations); and
  • they don’t attach disproportionate weight to the tool’s recommendations and can take into account relevant data and factors on which the decision was based.

Use case: tools used to provide ‘fit’ scores

An employer: 

  • requires candidates to complete:
    • a personality test;
    • a gamified behaviour or skills test; or
    • a written interview, such as with an AI chatbot; or
  • uses an AI tool to assess the candidate’s written application. 

These tools profile candidates. The hiring manager presets the desired personality types, behaviours or skills. The tool automatically calculates each candidate’s performance and produces an overall fit score. 

The tool automatically rejects candidates with an overall score below a minimum pass mark via a pre-scripted email. This decision constitutes ADM and is, therefore, within the scope of the ADM provisions.

The tool ranks candidates who exceed the minimum pass mark from highest to lowest. Hiring managers offer interviews to the highest-scoring candidates. The candidates who do not receive an interview are manually rejected by a hiring manager, who does not consider every application. This means there has been no meaningful human involvement. This decision also constitutes ADM and is within the scope of the ADM provisions.

When a hiring manager appoints or rejects those candidates who receive interviews based on a face-to-face interview, this decision is not solely automated.

Use case: mandatory screening questions

An employer requires candidates to answer mandatory screening questions as part of their application. These questions ask whether they meet essential job requirements, such as being eligible to work in the UK or having required qualifications or membership of certifying bodies. A human presets the questions and required responses.

The tool automatically rejects candidates who do not meet mandatory job requirements. This does not constitute ADM, because the tool is limited to taking an action predetermined by a human in response to a binary outcome. If the response is ‘no’, the tool rejects the candidate. If the response is ‘yes’, the tool progresses the candidate to the review team.