How this could lead to more discrimination in employment
As our computer software and hardware continually improves, employers are finding new ways to take advantage of these advancements. Instead of being relegated to increasing productivity, new technologies, like artificial intelligence, are taking on additional roles in the workplace. Applications include employee monitoring and hiring process assistance.
But using new technologies for decisions traditionally made by humans might have some growing pains, including inadvertent discrimination. Fortunately, civil rights enforcement agencies in the workplace like the U.S. Equal Employment Opportunity Commission (EEOC) and the Civil Rights Division of the U.S. Department of Justice (DOJ) anticipate this potential problem and have taken steps to address it.
How employers are using new technologies in the workplace
There are two main areas where employers use technology in ways that could lead to unlawful discrimination. First, it’s about monitoring employee productivity levels and measuring their performance. Second, it’s when making hiring decisions.
The use of software, artificial intelligence and algorithms to increase the efficiency of the hiring process might be the biggest example of how companies are trying to make the most of new technologies. But in this quest to save time, money and effort, some job applicants may find themselves unfairly and unlawfully disadvantaged.
How Advanced Technology Could Lead to Unintentional Discrimination in the Workplace
When it comes to employee monitoring (and ignoring the potential legal issues involved), there are situations where the goal of identifying who the most productive workers are could result in violations of the Americans with Disabilities Act of 1990. (ADA).
For example, an employer may have software that monitors an employee’s keystrokes or computer tasks performed over a period of time. But what if an employee has an ADA-recognized disability that makes typing or using a traditional mouse difficult?
To deal with this challenge, they might have access to a reasonable accommodation, such as voice recognition software or another type of input device. However, if the monitoring software does not take these adjustments into account, the employee with a disability could be unfairly noted by the monitoring software. This erroneous assessment could overlook the fact that the employee is fully capable of performing their job duties with reasonable accommodation. An employer aware of this scenario could avoid this situation by using another way of testing or scoring employee productivity levels. But the trick is to be aware that this type of scenario could arise.
As for technology leading to discrimination in the hiring process, the potential pitfalls for an unwary employer are numerous. Here are some examples :
- Software and algorithms that analyze candidate resumes for specific keywords.
- Facial and voice recognition technology that rates candidates based on the way they speak and their facial expressions.
- Ask potential employees to take a self-assessment test to determine their personality and whether they will be a good fit for a particular position.
- Use chatbots or virtual assistants to ask candidates basic questions about their qualifications.
- Require job applicants to pass a computer-based pre-employment test.
These seem innocuous, and for the most part, they are. So how could they lead to discrimination? In example 1, whoever develops the algorithms and software can unknowingly program the technology to give particular preferences to sexist words. This could give CVs of a particular sex or gender an advantage in the selection process.
In Example 2, the software and algorithms could place special emphasis on candidates who speak a certain way or use certain words. But those with accents, a unique way of speaking, or disabilities that affect speech could be penalized by the technology. This could lead to unlawful discrimination based on a person’s race, color, national origin or disability.
For example #3, a candidate with severe depression could be disqualified from a job because of their “personality” even if their answers are the result of an ADA-recognized mental impairment.
For example #4, imagine a chatbot that asks candidates yes or no questions about their career path. And one of the questions is whether they have ever been fired “for cause”. A candidate wants to answer honestly and err on the side of caution by answering “yes”. However, this response automatically results in their disqualification from the position. But the chatbot did not allow the job seeker to explain that the reason for his dismissal was later found to be illegal as it was due to the age of the job seeker.
In the fifth example, the criterion itself may not be discriminatory. But if it’s given in a way that puts some applicants at a disadvantage because of a protected trait, it could be illegal.
A hypothetical situation might be an aptitude test to be taken on a computer in the employer’s office that measures the candidate’s typing skills. But the job candidate is visually impaired. And although they can type on a regular keyboard, their typing skills are nowhere near as good as when using a braille keyboard.
What can employers do to avoid technological discrimination?
The biggest thing employers can do is educate themselves about potential problems with new technologies. And that despite the best intentions, discrimination can still happen.
In the majority of situations, employers who use technology do not seek to discriminate. But they might not know how the use of their technology results in the laying off of promising candidates and high performers due to protected characteristics.
To help increase employer awareness, various government agencies have published legal and technical advice. These are intended to advise employers on what to look out for when implementing new software, algorithms and artificial technologies in the workplace.
Two recent releases have come from the EEOC and the DOJ with The Americans with Disabilities Act and the use of software, algorithms and artificial intelligence to assess candidates and employees and algorithms, artificial intelligence and the discrimination of persons with disabilities in hiring, respectively. These focus on disability discrimination, but are probably just the beginning.
The US Department of Labor (DOL) will likely issue its own legal guidance for employers. Next, guidance will likely be given on how an employer’s use of technology could result in discrimination in violation of other federal laws, such as Title VII of the Civil Rights Act of 1964 (Title VII).
New technologies can provide new ways for employers to get work done. But if they’re not careful, it could lead to unintentional discrimination of employees and job applicants. Federal agencies responsible for enforcing civil rights in the workplace are already taking steps to help employers avoid these mistakes.