AI – the technology against discrimination in recruiting?

Authors: Lotte Bouwmans, Thomas Thiel, Flora Stein

AI is everywhere these days, much more than you might think. Just think: maybe you recently bought a new piece of clothing that was recommended to you online, or prepared a new recipe that you saw in a sponsored ad, or maybe Siri told you today what the weather was going to be? It is even possible that you were hired thanks to AI. It is seen as the technology of the future for a reason. But how is it possible for a system to know our preferences before we even think about it? In this article, you will learn more about AI itself and how it can influence the world of recruitment.

As mentioned earlier, AI is becoming more and more common these days. It can be found in phones, watches, facial recognition, but also in HR work life in departments such as recruitment (, 2021). AI in recruitment works like this: Advertisements for the job offer are placed via AI, so that only candidates, that could be already suitable for the job, see them. This works by using for example the history of online research. Next, a machine takes in all the incoming data (applications) and selects on the preferences entered by the recruiter (Dilmegani, 2022). If the candidate has any questions, they can be answered by a chatbot. After the resume screening, the filtered candidates receive an auto-Email that is still personalized for the individual candidate. The candidate can set up an interview appointment with an interview software. Even the interviews can be replaced by interview videos that are screened by AI. A software can analyze the candidate’s body language and choice of words. The only work that is left for HR recruiters is leading the final interviews (Dilmegani, 2022). The purpose of using AI is to improve the organization’s recruitment process. The time spent on repetitive, high volume tasks can easily be reduced through this kind of automation.

According to Nunez (2021), recent studies show that 99% of Fortune 500 companies already use some form of AI tools in their recruitment process. It is expected that this number of companies will only increase in the coming years.So AI in recruitment is already being used in many companies. Usually, not all steps are taken over by artificial intelligence. Audiobook streaming service Audible, for example, uses Ad Placement software to primarily reach qualified candidates and reduce its advertising budget. L’oreal, for example, uses chatbots to answer questions from candidates. The initial screening is replaced by an online quiz for the candidate. The planning of the job interview is also done by software. Another example is Hilton, which also uses interview software. The questions are programmed to increase efficiency and save time.

“I myself worked a lot with recruitment during my HR internship; the department was completely entrusted to me. The company I worked for was a start-up, so many of the online programmes/systems did not yet exist or did not work properly. Among other things, this made the application process very time-consuming. After someone had applied via one of the online platforms, a message would automatically appear in the mail. This application then had to be manually assessed by me and processed in Excel. I was the one who conducted the first interview and forwarded or rejected applicants to the manager. This meant that I had full responsibility for determining who was or was not suitable for a job.

In my opinion, it is very risky to put so much responsibility on one person. (Un)conscious personal preferences can so easily play a role in the selection of candidates. It is also the case that far too much time is spent manually processing all the applicants. Time that could be used much more usefully, for example to attract new applicants.” -Lotte Bouwmans, April 2022

Most companies reason their use of AI recruitment with increased efficiency and decreased costs. When using AI for job advertising placements, the advertising budget is used more efficiently. Otherwise, ad money is spent unnecessarily for an unsuitable target group (Nunez, 2021). Resume screening by humans takes much more time than automatically. With AI more applicants can be scanned to increase the chance of finding suitable candidates. The hiring costs are also decreased this way, since less HR personnel is needed. This also leads to a higher value of human work (Author, 2021). AI recruitment also has advantages for the candidates because the application process happens faster. After AI resume screening, applicants get responses much faster. This way, suitable and qualified candidates are more likely to stay onboard during the application process (Dilmegani, 2022). Another topic in the recruitment process is discrimination. In contrast to recruitment by HR personnel, AI recruitment is supposed to be unbiased. The filters for the resume screening are qualifications and skills that are needed for the job. Appearance, race or gender are left out, so that candidates are only chosen by their qualification (Author, 2021). The filters can also be set in order to reach certain diversity goals (Nunez, 2021).

At first sight, AI recruitment seems like the perfect tool for the future. Like most inventions, there are also risks that come along implementing a machine as HR personnel. Firstly, the amount of jobs in the HR apartment decreases, since human work is replaced by machines. This also leads to a recruitment process that is less personal. Providers of AI recruitment tools advertise with efficiency regarding time and budget. But does the recruitment process have to be only efficient? Work is for most people a huge part of their life, which is why it is also important that working teams fit well and employees feel comfortable at their job. When using AI recruitment, the emotional aspects that are typical for human work are completely left out. Another topic that needs to be considered when working with algorithms is data protection. AI tools collect a huge amount of data from the candidates, which also means responsibility of the recruiting company to protect the data well. Critics also say that the quality of the collected data is lower than the information a human would find out about a candidate, e.g. in an assessment center. As mentioned before, AI recruiting works with algorithms. These work by being fed with data. You can simply say that an algorithm learns from humans. This is why it is not guaranteed that AI recruitment is unbiased. It always depends on the filters and criteria the recruiting company sets and those can be discriminating, too. Furthermore, it is hard to evaluate how well the AI recruitment works, since the rejected candidates are often not traceable. In general, the whole recruitment process is less transparent because it is hard to trace the work of algorithms (Goretzko & Israel, 2022).

As we can see, AI recruitment is a discussed topic on which a lot of research still needs to be done. At the moment most companies do not use AI for the whole recruitment process, but it replaces a few steps. This is probably the best solution for now, since algorithm based hiring also brings a few risks. So, getting hired by a machine is the recruitment for the future, since it still has to be optimized. The authors found that the most disadvantages of AI in recruitment are things, that the AI industry is currently working on.


Author, G. (2021, February 17). The Role of Artificial Intelligence (AI) in Recruitment.Blog. Retrieved April 26, 2022, from

Dilmegani, C. (2022, April 4). Recruiting AI in 2022: Guide to augmenting the hiring team. AIMultiple. Retrieved April 25, 2022, from (2021, February 23). AI in daily life. Retrieved April 25, 2022, from,Cortana%20help%20are%20forms%20of%20everyday%20artificial%20intelligence.?msclkid=8bdddd5ac48b11ecb9129353de05d72e

Goretzko, D., & Israel, L. S. F. (2022). Pitfalls of Machine Learning-Based Personnel Selection. Journal of Personnel Psychology, 21(1).

Nunez, A. (2021, December 13). 3 Companies That Are Using AI Tools for Recruiting. Recruiting: AI-Enabled Recruitment Platform | Pandologic. Retrieved April 25, 2022, from

Leave a Comment