AI in recruitment: friend or foe?
DCA has partnered with Hudson RPO and Monash University to research the impact of unconscious bias on recruitment and selection decisions using AI and offer interventions to minimise or remove the influence of unconscious bias in recruitment.
Artificial Intelligence (AI) has incredible potential to positively influence the diversity of new hires as well as reduce bias in recruitment. But, like any tool, it’s not failproof. How effective is AI at reducing bias? How can human recruiters work with AI to improve diversity outcomes? And how much technical expertise is needed to select and use AI properly?
In this first stage of the three year study, AI: Neither Friend nor Foe for D&I, we asked hiring professionals, AI developers, academics and industry experts to share their insights into the use of AI tools in recruitment and their impact on diverse people.
What we heard was that while AI can be an efficient, convenient, supportive, and objective tool that is capable of analysing bias in recruitment. For AI to support inclusive recruitment, there needs to be more support for people – recruiters, HR professionals and developers – on how to use these tools to eliminate rather than amplify bias. This means developers applying a D&I lens during the design and testing of these tools, and recruiters and HR professionals applying a D&I lens when AI tools are being deployed.
WHAT YOUR ORGANISATION CAN DO
USE AI TO ELIMINATE NOT AMPLIFY BIAS:
1. Ask for evidence that your AI tool is bias-free before deploying it.
2. Consider job design through a D&I lens before deploying AI.
3. Make sure you incorporate vendor advice about training recruiters and HR professionals in using AI tools.
4. Use AI to identify bias in new hire patterns.
5. Use AI recommendations as just one data point.
6. Ensure that staff using AI tools have adequate training.
Want to Use Our Research?
Materials contained in this document are © Copyright of DCA Ltd and Monash University, 2022. If you wish to use any content contained in this report, please contact DCA at [email protected], to seek consent.
Where you wish to refer to our research, it must be correctly attributed.
- Formal attribution to DCA & Monash University is required where references to DCA research material are in a written format.
- Citing DCA & Monash University as a source will suffice where the reference is made in a verbal format.
Suggested citation: Monash University/Diversity Council Australia, AI in recruitment: friend or foe?, Sydney, Diversity Council Australia, 2022.