AI bias denotes any way that AI and data analytics tools can perpetuate or amplify bias. An example of an ethical implication is the use of artificial intelligence in the hiring process of companies. Since the introduction of AI in the hiring process, recruitment is more efficiently processed, cost-effective, and better targeted at processing a huge volume of resumes (Parikh, 2021). However, this way of processing undermines fairness and inadvertently dis-seminate bias (Jobin et al., 2019). The bias is created by the data, if the data set is not representative and diverse, then the results will be clouded. The main problem with this historic data is that the industry portrays an ideal candidate with a certain degree or cultural background, therefore the AI build-up on this data is inherently biased. Based on the historic data the ‘perfect’ candidate would consist of a white male with an ivy league background. While this is far from reality.
Organizations can minimize the bias in AI by updating the data set with a wide range of employees, with different degrees, cultural backgrounds, and work experiences, because there is no perfect resume to match a candidate with the company. Also, it is important to always have a second opinion from actual HR employees. Human interference can consequently undermine the bias created by AI. The biggest drawback is the lack of human judgment (Parikh, 2021). AI-based hiring may not serve its purpose if a company intends to diversify its workforce. Candidates with atypical work experience could be the best match for the company in regard to their work ethics, character, and interest resulting in a higher employee retention rate. Whilst AI will perhaps miss these potential candidates. The problem is recognized worldwide, and even leads governments to take initiative to adopt rules and regulations. The European Union has proposed a regulatory framework for AI, which can help identify the bias portrayed by AI in the hiring process (Lohr, 2021).
What do you think? Is the efficiency of using AI far greater than the risk of amplifying bias in the hiring procedure?
References:
Jobin, A., Ienca, M. and Vayena, E., 2019. The global landscape of AI ethics guidelines. Nature Machine Intelligence, 1(9), pp.389-399
Lohr, S. (2021, December 8). Group backed by top companies moves to combat A.I. bias in hiring. The New York Times. Retrieved 7 October 2022, from https://www.nytimes.com/2021/12/08/technology/data-trust-alliance-ai-hiring-bias.html
Parikh, N. (2021, October 14). Understanding bias in AI-enabled hiring. Forbes. Retrieved 7 October 2022, from https://www.forbes.com/sites/forbeshumanresourcescouncil/2021/10/14/understanding-bias-in-ai-enabled-hiring/?sh=43d455d77b96
Hi Lubna, thanks for this interesting article! I think the usage of AI can really harm hiring procedures. Sometimes a person’s education and previous experience are the perfect match for a certain position, but maybe the person’s characteristics do not match with the company culture. This aspect is essential, because the person itself must fit within the team and company he/she will be working in. I think AI will never be able to make this choice better than humans do, as it can be a certain ‘feeling’ that the hirer has with the participant. Therefore I agree with you that there always have to be a second round with an (HR) employee.
Interesting article Lubna! I agree that AI can increase biases and racism during the recruiting process. I would suggest that AI and human capital are combined to get a balance and reduce racism. For example, AI could focus on words such as “university”, “skills”, “honours” etc., whereas recruiters focus on a more in-depth analysis of resumes and cover letters. I think you have addressed a very important issue of AI, well done!