Published Article

Automating HR Functions in Education: AI in Schools

November 16, 2020

The last decade has brought major changes and challenges to both higher and secondary education. Schools constantly find themselves contending with traditional issues such as classroom size, student engagement, access and equality. However, underlying each of these issues is the recruitment, retention, training and hiring of high performing educators. With reduced government funding compelling many school systems to shift their resources, and the new demands brought on by COVID-19, human resource departments have had to adapt to support a changing workforce. One of the primary questions arising from this shift is, how can human resource professionals adapt to attract and retain the best talent? Some schools are turning to Artificial Intelligence (AI) for the answer.

AI has become an increasingly common and vital technology used in homes and businesses. Some of the most common examples of AI today include Google Home, Siri and Amazon Alexa. While AI has existed in some form for decades, in the era of the coronavirus (COVID-19) pandemic, use of AI in the workplace is increasing, as employers seek ways to limit exposure to and prevent the spread of the virus.

In the education context, AI technologies open up new ways for students to interact, particularly granting access to education for students with special needs and disabilities. AI can also be used to assess the academic areas in which students may face problems, which helps in creating customized learning paths. AI technology can also be used to optimize grading and communicating with students, which are often time-consuming administrative tasks for teachers. AI technology is also being used by some school districts facing teacher shortages. During the 2016-17 school year, 71 schools in Mississippi used the state’s online learning platforms according to the Mississippi Department of Education. In 2018, the number grew to 106 schools. According to data from research firm Technavio, it is expected that artificial intelligence in U.S. education will grow by 48% through 2022. With the continued rise of AI, this article highlights three key legal issues that school administrators should be mindful of from a human resources perspective.

Implicit Bias and Discrimination

Using AI to recruit high performing educators shortens the hiring window, and also offers the promise of reducing or eliminating the unconscious bias that can inform hiring decisions. Whether conscious or not, bias continues to affect decision making, which adversely affects people of color, according to a study from Harvard’s Radcliffe Institute. AI tools may be programed to disregard a job applicant’s name, gender, race, or ethnicity and focus on other factors.

Employers, however, should be mindful that AI will not shield them from discrimination claims. Employers must still comply with all federal, state, and local anti-discrimination laws in all aspects of the employment relationship, including the pre-employment screening and interview process. For example, pre-programed robotic systems that operate on voice commands may have an inherent bias against applicants that do not speak English. Employers should program advanced systems that take into account various languages and national origins of employees. Similarly, an algorithm that sends advertisements for managerial positions exclusively to men and administrative assistant positions exclusively to women, based on past applications or hiring patterns, may not necessarily be designed with the intention of creating categories based on identity characteristics. According to 2019 research from Arizona State Law School, however, once it does so, it almost always involves intrinsically offensive practices that risk reinforcing the importance of gender in path-dependent ways. Artificial discrimination has the ability to impact terminations as well. The Age Discrimination in Employment Act (the ADEA) prevents discrimination against individuals over the age of 40. 29 U.S.C. 14 Section 623. The use of robotics could present unique challenges to employees and applicants who may have limited experience with or exposure to new technologies. Employers should work hard to ensure that biases concerning older individuals and their use of technology are not driving hiring decisions.

The Wisconsin Supreme Court has provided a two-part road map on how courts will likely permit and support the growing use of AI in human resources decision-making. see State v. Loomis, 371 Wis. 2d 235 (2016). First, employers should keep a human review as part of the AI-enabled HR decision-making process. Second, the employer should disclose that AI is being used in the decision-making process.

Collective Bargaining and AI

The growth of unions in higher education has generally stalled at about 375,000 faculty—21% of private colleges and 35% of public colleges. However, unionization in secondary education is undoubtedly on the rise. This is in large part due to the National Labor Relations Board’s (NLRB) decision from earlier this year, wherein the NLRB held that it can assert jurisdiction over charter schools. See KIPP Academy Charter School, 368 NLRB No. 48 (2020).

The use of AI in the workplace presents additional challenges for employers with a unionized workforce. When a union is certified by the NLRB or voluntarily recognized by an employer as the representative of employees in a bargaining unit, the union becomes the exclusive bargaining agent for all employees in the unit, regardless of individual employees’ membership in the union. The NLRB has long held that technological changes that significantly affect an employer’s unionized workforce are a mandatory subject of bargaining. Use of artificial intelligence may fit the bill. See Omaha Typographical Union, No. 190 v. NLRB, 545 F.2d 1138 (8th Cir. 1976). Worker displacement and layoffs, whether caused by AI-powered technology or otherwise, are also a mandatory bargaining subject. The decision to implement new technology or workplace automation powered by AI may also be subject to mandatory bargaining under certain circumstances. See First National Maintenance v. NLRB, 452 U.S. 666 (676-77) (1981); Fibreboard Paper Products, 379 U.S. 203, 214-15 (1964).

Employers should ensure that any collective bargaining agreement contains a robust “management rights” clause that reserves the employer’s right to make operational changes, as well as a “no strike” clause that will prohibit employees from striking over such changes during the term of the agreement.

AI and Wage-and-Hour Issues

Employers that incorporate AI into the workplace must also consider potential compliance issues under federal and state wage and hour laws. Pursuant to the Fair Labor Standards Act (FLSA), unless exempt, employees who work over 40 hours in a workweek must receive overtime pay at a rate not less than time and one-half their regular rate of pay. There are several exemptions to the FLSA’s overtime provisions including the executive exemption. 29 C.F.R. part 541. To qualify for the executive exemption, an employee must satisfy both the salary test and duties test.

To satisfy the duties test, an employee must:

• Manage the enterprise, or manage a customarily recognized department or subdivision of the enterprise;
• Customarily and regularly direct the work of at least two or more other full -time employees or their equivalent; and
• Have the authority to hire or fire other employees, or the employee’s suggestions and recommendations as to the hiring, firing, advancement, promotion or any other change of status of other employees must be given particular weight.

If an employee is not exempt, however, all work time must be paid even if off the clock. To the extent AI causes extended work, proper pay protocols must capture all time worked.

AI may result in the redefinition of the duties of some employees. Employers who automate certain job duties or positions must ensure that workers continue to manage two or more employees. If the employee no longer qualifies for the exemption, the employer should reclassify the individual as a nonexempt employee, pursuant to the Fair Labor Standards Act.

Reprinted with permission from the November 13, 2020 issue of The Legal Intelligencer. © 2020 ALM Media Properties, LLC. Further duplication without permission is prohibited.  All rights reserved.