Anokhi Manchanda
Associate Editor
Loyola University Chicago School of Law, JD 2025
Around 66% of Americans would not want to apply for a job if artificial intelligence was used to make hiring decisions. As AI has become ubiquitous, it has started to emerge as a prominent tool used in hiring processes. Now, about 83% of employers use some form of AI in their hiring practices.
Usage and Advantages of AI tools in the hiring process
The broad range of AI talent search tools usually are used to find applications with specific keywords. They are also used to find workers with certain types of experiences or skills. The data on these tools are limited and, because many were formed without scientific methods or proper research, we cannot expect the tools to find candidates faultlessly. Although, the AI tools have helped human resource and hiring departments search through resumes much faster than previously. The hiring process is more efficient, and it is easier to find the best suited candidate for the search. Therefore, the process often becomes cheaper as well.
Arguments against using AI in hiring
However, there are many concerns with how the AI tools are “trained.” For example, if a company uses current/past employees’ resumes to create data for the tools to search for in potential candidates, the algorithm may produce biased results based off the company’s existing biases and past hiring practices. A professor of organizational behavior and public policy at Carnegie Mellon University stated that employers are risk-averse and therefore find many reasons to disqualify people. For example, AI tools may exclude a resume that shows gaps in a person’s career if a company wants to disqualify a candidate for long periods of unemployment. The tools would reflect the existing biases and may then exclude someone who took time off after giving birth or because they were taking care of a loved one. A recent example is a tech company that used existing employees’ resumes to train the AI hiring tools which subsequently caused discrimination against applicants who were women.
Legal implications
There are many avenues that people who are discriminated against with this technology can defend their rights. One course of action is through civil rights laws, which bar discrimination in person and online. Although it may be hard to identify when you are discriminated against with this technology, other organizations will have to help mitigate these issues. For example, one woman applied for a job position twice using different birth dates and later realized that the position had an unlawful age cutoff. The Equal Employment Opportunity Commission (EEOC) filed suit on her behalf. Additionally, employers are not allowed to ask about attributes such as age, gender, marital status, etc. Yet, these AI tools can gather information regarding those factors without the consent of the applicants. As per the EEOC’s instructions, if it appears that employment-related rights were violated in the application process, one should file a charge of discrimination with the EEOC so they can investigate and remediate the issue.
Recommendations for the future
The aforementioned information is quite alarming, given that AI technology used in hiring can perpetuate harmful biases that we are working to prevent. Moving forward, there should be transparency on when and how AI is used in the hiring process throughout the employment as well. Other forums suggest that people should use AI to create and enhance their resumes. A video interview platform called HireVue removed the facial analysis component from its hiring process due to concerns of inappropriate uses of AI in the hiring process. Additionally, many professionals agree that audits should be conducted to ensure that the AI used by companies are being trained to avoid discrimination. A bill was proposed in New York City to require vendors to audit their tools to ensure there are no discrimination concerns. Yet another idea is that AI hiring tools should be trained to increase the diversity of job candidates.
After conducting this research, I think that companies should stop using this artificial intelligence technology in hiring processes (or use it sparingly) to avoid further discrimination. There should be more regulation regarding audits or some tests put in place that the AI tools must pass so that quality candidates are not illegally or unnecessarily excluded from applicant pools. Making the hiring process faster or cheaper could lead to a variety of short and long term legal implications for companies, like hiring discrimination lawsuits, and an eventual lack of corporate diversity. Companies should want to avoid these for ethical and practical reasons. Because it is possible to train AI software to include diverse candidates even if the company has not historically been diverse, laws should be put in place to ensure inclusion and ban this new type of discrimination.