AI Has a Gender Diversity Crisis, and Here’s Why HR Should Care

Despite an industry-wide push towards diversity & inclusion (D&I), female representation in the AI sector continues to be below average. And this has a definite impact on how your AI-based HR systems approach decision-making. In this article, we discuss:

• Insights from the recent AI Now report and what they mean for HR 
• 3 HR areas that the lack of gender diversity in AI could affect
• Strategies that could help you foster a fair and equitable workplace while gaining from AI

 

HR is increasingly using artificial intelligence (AI) to simplify tasks in different areas, ranging from recruitment and onboarding to workforce management and payroll. AI systems can process large volumes of data in a fraction of the time, drawing actionable insights to guide HR. What’s more, AI is compatible with both structured and unstructured data. This means that you could potentially extract information from a candidate’s social media or pre-recorded video to aid in the interview process, and this could lead to a certain amount of bias.

But, are the insights provided by AI always reliable and objective? Recent research has surfaced showing alarming figures about diversity in the AI sector.

A report by the AI Now InstituteOpens a new window , New York University, revealed a glaring diversity gap in the field of artificial intelligence, which will affect how these systems approach data and make decisions.

There is a palpable “trickle-down” effect where the lack of AI diversity could lead to bias-prone HR systems, further perpetuating its impact on your workforce. Let’s look at some of the key findings of the AI diversity report in greater detail. 

Learn More: Why Is Artificial Intelligence Biased Against Women?Opens a new window

What AI Diversity Statistics Mean for HR

The tech industry has gained a reputation for being biased against female candidates and workers from minority groups. While the industry average is showing incremental improvement, the case of diversity in artificial intelligence demands closer attention:

• Approximately 25% of the tech workforce in the U.S. comprises women Opens a new window – and the number is even lower for AI. Only 18% of authors at leading AI conferences are women, with over 80% of positions still dominated by male professors.
• Even progressive companies such as Facebook and Google are lagging when it comes to AI diversity. Only 15% of AI research staff at Facebook comprises women, and the number is even lower at Google (10%).
• There is significant variance among women of different ethnic backgrounds with Caucasian women still preferred over those of other minorities. 

 

This implies that the people designing and building your AI-based HR systems could bring a biased perspective to product development. As a result, these ideas are perpetuated across workplaces that use these tools.

Learn More: How Employers Are Using AI to Stop Bias in HiringOpens a new window

3 Aspects of HR Affected by Artificial Intelligence

As AI transforms how you manage your workforce and complete day-to-day tasks, it is essential to keep an eye on the possibility of bias at every level. AI requires a balance with human intelligenceOpens a new window , but this human intelligence needs gender diversity to create a solution that can consider multiple factors when making important decisions.

This covers three critical HR functions: 

1. An AI chatbot that screens your candidates 

Candidate screeningOpens a new window is a popular AI use case, implementing a chatbot to assess basic candidate information for first-level screening. Based on data such as experience, location, career goals, and educational qualification, the chatbot creates an initial shortlist that can be followed up by a human HR executive.

However, if a bot is programmed with a degree of bias, it will screen candidates based on discriminatory parameters. For instance, since there is a shortage of women in tech, it may shortlist more male candidates than female candidates for a tech-based role. And the built-in machine learning algorithm will only strengthen the bias over time. 

2. An AI-based automated workforce scheduler 

Employee schedules can be dynamically set and configured using AI. This helps to synchronize work schedules as per every stakeholder’s convenience while enabling higher productivity. In case bias creeps into the development stage of the scheduler, you could risk over-burdening certain employee segments while leaving other resources relatively idle. For instance, the solution could bench women and schedule more jobs for men, especially in jobs that are dominated by men.

An automated scheduler may not always take individual desires into account, choosing to follow group-based rules instead. 

3. Performance management using AI 

Employee evaluations and appraisals are other areas where AI can give HR greater visibility into employee performance and achievements. It can process data from a variety of sources and streamline the insights generation process. However, bias can skew how the AI looks at “good” or “bad” performance.

For example, a female employee who is a new mother and may require more time off than usual may be rated as being demotivated or uninterested in the job.

These areas can also result in the lack of women in leadership roles, as their progress is hampered by human bias reinforced in AI solutions that measure their performance in the workplace.

Learn More: Can Artificial Intelligence Eliminate Bias in Hiring?Opens a new window

How HR Can Leverage AI Responsibly: The Way Forward

Despite the poor state of gender diversity in AI, the core technology could completely change HR efficiency. A future-focused organization cannot afford to hold onto legacy, manual HR processes. By staying aware of bias risks and the need for greater AI diversity, AI can be used in HROpens a new window to promote inclusive and fair workplace practices. Here are a few concrete strategies that will pave the forward. 

1. Understand precisely when hiring algorithms should enter the recruiting processOpens a new window . By applying AI to a gender-inclusive and diverse candidate pool, you can cut down on the impact of bias and still gain from the decision-making capabilities of the AI engine.

2. Adopt a skill-focused culture across the organization. This goes beyond AI-led recruitment. Attention to proven skill sets (both hard and soft) at every step, from hiring to rewards and performance management, will ensure that your employees can enjoy a fair and equitable workplace. 

3. Upskill your workforce with the knowledge of how AI works. This will allow employees to spot any instance of bias due to the lack of AI diversity and promptly connect with your software vendor to address this. Without internal capabilities, issues like this could continue to be overlooked, perpetuating the adverse effects of low diversity in the AI sector. 

Technologies such as AI and automation are poised to become the driving force for workplace transformation. Therefore, AI diversity must be tackled head-on.

Technology vendors must proactively build a diverse workforce to build solutions that enable the creation of diverse workforces in other organizations. But the onus is also on HR to demand more equitable and socially responsible tools to make unbiased decisions.

By acutely understanding the ethics of AI in HROpens a new window and the impact of AI as well as its many benefits (and risks), you can steer your workplace towards a positive, more efficient future. 

How has AI impacted your workplace? Share your experiences, positive or negative, with us on FacebookOpens a new window , LinkedInOpens a new window , or TwitterOpens a new window . We are always listening!