When AI Meets DEI: Is HR Tech Helping or Hurting Workplace Equity?

AI can remove bias – or reinforce it. How can HR leaders ensure AI-driven decisions create more opportunities, not fewer? 

Introduction – The Opportunity and Risk of AI in HR 

In the last three years Artificial intelligence has rapidly transformed HR, revolutionizing processes like hiring, onboarding, performance management, employee engagement, and talent development. Companies are turning to AI to streamline decision-making, enhance efficiency, provide personalized experiences, and improve workforce planning. The promise of AI in HR is compelling – it can analyze vast amounts of data faster than any person can, identify patterns in the data that might otherwise go unnoticed, and even potentially reduce bias in hiring and promotions by removing human subjectivity. As organizations strive to assess and race to adopt AI-powered solutions, they see an opportunity to create more equitable, data-driven decision-making frameworks that benefit both employees and employers. 

However, AI models are only as objective and unbiased as the data from which they were trained. Many AI models used for HR are trained on historical workforce data, which often reflects longstanding biases present in an organization’s historical hiring, promotion, and performance evaluation processes. This means that rather than eliminating bias, AI can sometimes reinforce it – scaling discrimination in the organization instead of helping to reduce it. 

You may recall that in 2014, Amazon initiated an experiment at its Seattle headquarters to use artificial intelligence for streamlining the hiring process. The AI system was designed to review resumes and recommend top candidates. However, by 2018, the experiment was discontinued after it was discovered that the AI exhibited significant gender bias, favoring male candidates over female ones. This bias stemmed from the AI being trained on resumes submitted to Amazon over a decade, which predominantly came from male applicants. Consequently, the AI system downgraded resumes from women, leading to its eventual scrapping. This incident highlighted the challenges and limitations of using AI in hiring processes. 

As the Amazon example shows, even well-intentioned organizations may inadvertently deploy AI systems that disadvantage underrepresented groups, leading to unintended consequences in diversity, equity, and inclusion (DEI) efforts. Without careful oversight, AI may become a barrier to the very inclusivity that HR leaders strive to foster. 

Real-world examples of AI bias in HR are already emerging. As the Amazon example showed, automated resume-screening tools have in the past tended to favor male candidates over female candidates in tech hiring, facial recognition software can struggle with accuracy across different ethnicities, and AI-driven performance assessments may unfairly penalize employees based on language patterns or communication styles. At the same time, AI is being used more intentionally and in ways that could advance DEI – such as identifying pay disparities, helping organizations measure inclusion efforts, or removing biased language from job descriptions. The key question is whether companies are leveraging AI responsibly and with a deep understanding of its impact on workplace equity or are they simply deferring and accepting an AI’s recommendations. 

As AI continues to reshape HR, the challenge for leaders is not whether to adopt these technologies but how to do so in a way that supports, rather than hinders, DEI goals. HR teams must ensure that AI tools are rigorously tested for bias, used transparently, and integrated with human oversight. Organizations that proactively address AI’s impact on inclusion will be better positioned to build truly equitable workplaces, while those that ignore these concerns risk perpetuating the very biases they seek to eliminate. The future of AI in HR depends not just on technological advancements but on the ethical responsibility of HR leaders to guide its use in a way that strengthens diversity, equity, and inclusion rather than undermining it. Let’s explore these issues in more detail, first by examining how AI may influence DEI initiatives and progress. 

The Inclusion Challenge – Understanding AI’s Potential Impact on DEI 

Many organizations are now racing to implement AI solutions, but fewer fully assess the DEI implications of adopting these new technologies. Recent data from Paychex shows that 72% of small and mid-size businesses have expressed a somewhat positive to a very positive attitude toward AI and 82% viewing AI technology as helpful to their business. It is clear that HR leaders have shown the willingness to embrace AI and are eager to implement AI-driven HR solutions, drawn by the promise of efficiency, cost savings, and improved decision-making.  

And that willingness and enthusiasm should not be surprising given the remarkable power and potential of these technologies. AI-powered tools can analyze massive datasets, automate routine HR tasks, and even identify patterns in hiring and employee performance data that might otherwise go unnoticed. However, in the rush to adopt these technologies, companies often fail to fully assess how AI impacts diversity, equity, and inclusion. While AI has the potential to remove human biases, it also has the capacity to replicate and even exacerbate systemic inequities—especially when built on biased historical data. Without a thoughtful and intentional approach, organizations may unknowingly introduce new barriers to workplace inclusion. 

AI systems learn from existing datasets, which means that if historical HR data contains bias, or is simply “messy”, the AI will likely perpetuate it. For example, if a company’s past promotion decisions have favored certain demographics over others, an AI-driven talent assessment tool trained on that data may reinforce those same patterns rather than correcting them. And as we have seen in recruiting and screening processes, some AI models may downgrade candidates with non-traditional career paths, penalize job seekers from underrepresented backgrounds, or prioritize candidates who closely resemble past hires. These unintended consequences highlight the reality that AI is not inherently neutral – it mirrors and amplifies the biases present in the data it processes. 

Beyond the risk of biased decision-making, organizations must also consider the legal and ethical implications of using AI in HR. Governments and regulatory bodies are increasingly scrutinizing AI-driven hiring tools, and lawsuits over algorithmic bias are on the rise.  

In recent years, various legislative bodies have enacted laws and regulations to govern the use of AI technology in hiring processes, aiming to ensure fairness, transparency, and accountability. These regulations address concerns about bias, discrimination, and privacy, mandating measures such as bias audits, impact assessments, and transparency in AI decision-making. The goal is to prevent discriminatory outcomes and build public trust in AI technologies.  For example, New York City passed a law in 2023 requiring employers to conduct annual bias audits of their AI hiring tools and disclose the results to job applicants. This law is enforced by the city’s Department of Consumer and Worker Protection, with penalties for non-compliance including fines up to $1,500 per violation. 

Similarly, Illinois enacted the Artificial Intelligence Video Interview Act in 2020, which mandates that employers using AI to analyze video interviews must inform applicants, obtain their consent, and explain how the AI works. Violations of this act can result in civil penalties. In Maryland, a 2024 law requires employers to conduct impact assessments of their AI hiring tools to ensure they do not disproportionately affect protected groups. The Maryland Department of Labor enforces this regulation, with potential penalties including fines and mandatory corrective actions. 

Companies that fail to test and monitor their AI systems for fairness risk not only reputational damage but also potential legal consequences. Ethical concerns also come into play when employees and job candidates are subject to opaque AI-driven decisions without transparency or recourse. HR leaders must recognize that while AI can enhance DEI efforts when used correctly, it can also expose their organizations to significant risks if left unchecked. 

Where AI is Impacting HR: Key Areas to Watch for DEI Impact 

Recruiting and Hiring 

One of the most widely adopted uses of AI in HR is in recruiting and hiring, where AI-powered tools are used to source profiles, engage with candidates, screen résumés, schedule interviews, assess candidates, conduct video interviews and more. While these technologies promise to streamline the hiring process and reduce human bias, they can also introduce new forms of discrimination if not carefully managed. We’ve discussed how AI-driven résumé screening systems, for example, often rely on past hiring data to identify patterns in “successful” candidates, which can introduce bias if an organization has historically hired more men than women in a particular field for example. Similarly, AI-powered video interview platforms analyze candidates’ facial expressions, tone of voice, and speech patterns to assess traits like confidence or leadership potential. However, these systems can be problematic for individuals from diverse linguistic and cultural backgrounds, as well as candidates with disabilities, who may communicate differently than the AI expects. Without proper oversight, AI can become an invisible gatekeeper that filters out diverse talent rather than fostering inclusivity. 

Performance and Talent Management 

AI is also reshaping performance management and promotion decisions, with predictive analytics and sentiment analysis tools becoming increasingly common in evaluating employees. Many organizations use AI to assess productivity, analyze communication patterns, and even predict which employees are likely to succeed in leadership roles. Goal definition, goal writing, and progress made against objectives are also more likely to be AI-assisted or augmented. Newer solutions can tap into a year’s worth of peer feedback and recognition data to better inform managers at performance evaluation time. And in truth, HCM technology providers are just tapping into the potential of AI-assisted talent management, which if done carefully, could significantly improve processes that have long been a source of frustration for HR, management, and employees. 

However, these tools can reinforce workplace inequalities if they are not designed with DEI considerations in mind. For instance, AI-driven performance assessments may prioritize specific work behaviors – such as assertive or dominant communication styles and approaches or always-on availability – that align more closely with traditionally male-oriented workplace cultures and disadvantage employees with different working styles. Likewise, promotion algorithms trained on historical data may favor employees who fit the profile of past leaders, which can exclude women, people of color, and other underrepresented groups from leadership pipelines. If AI models do not account for systemic barriers that have historically limited access to advancement opportunities, they risk perpetuating an exclusionary status quo rather than fostering a more diverse leadership landscape. 

AI in Employee Engagement & Retention 

AI is increasingly being used to monitor employee engagement, assess employee sentiment and organizational culture, recommend career development opportunities, suggest mentors, and direct learning and skills development. AI tools are also being leveraged to help employees manage their financial health, understand their compensation, manage retirement contributions, and choose the most appropriate benefits enrollments based on their personal situation and goals. While these applications have the potential to enhance employee experience, they can also reinforce existing disparities in engagement and retention.  

For example, AI-driven career pathing tools analyze historical promotion data to suggest internal mobility opportunities for employees. If past promotions have favored certain groups, the AI may overlook high-potential employees from underrepresented backgrounds, limiting their career growth. Similarly, AI-driven learning and development platforms recommend training programs based on previous participation patterns, which may unintentionally reinforce existing skill gaps between different demographic groups. Additionally, sentiment analysis tools that monitor employee communication on platforms like Slack or email can misinterpret cultural and linguistic differences, leading to skewed assessments of employee morale or productivity. These can also potentially negatively impact employees who prefer more direct and in-person methods of collaboration and communication. Without careful oversight, these AI-driven insights could create a workplace where some employees are systematically overlooked for development, engagement, or leadership opportunities.  

How HR Leaders Can Use AI Without Undermining DEI 

As AI becomes more deeply embedded in HR processes, organizations must take proactive steps to ensure that these technologies support rather than hinder DEI efforts. HR leaders cannot afford to take AI-driven recommendations at face value, nor should they assume that AI is inherently neutral. Instead, they must actively audit, oversee, and refine AI systems to ensure fair and equitable outcomes. The following strategies can help HR teams leverage AI responsibly while safeguarding DEI in the workplace. 

1. Audit Your AI Systems for Bias

Before deploying AI-powered HR tools, organizations must conduct thorough audits to assess potential biases in their AI models. Bias audits involve testing AI outcomes across different demographic groups to identify disparities in hiring decisions, performance evaluations, promotions, and other HR functions. For instance, if an AI-powered résumé screening tool is disproportionately filtering out women or minority candidates, the organization must adjust its algorithms or training data to correct the imbalance. 

One example of this approach is how LinkedIn tackled bias in its AI-powered recruiting tool. The platform found that its algorithm was favoring male candidates for certain tech jobs because the underlying data reflected historic gender disparities in the industry. By reconfiguring the system to focus on skills and competencies rather than past hiring patterns, LinkedIn was able to reduce bias in its recommendations. HR leaders should work closely with AI vendors to demand transparency in their algorithms and ensure that bias testing and industry accepted auditing is an ongoing practice rather than a one-time assessment. 

2. Use AI as a Complement to Human Oversight

While AI can analyze vast amounts of data and detect patterns beyond human capability, it should never be used as the sole decision-maker in HR processes. AI is best utilized as a decision-support tool rather than a decision-making authority. Critical HR functions such as hiring, promotions, and performance evaluations should always include a layer of human review to ensure fairness and context. 

Balancing AI technology with the human element in HR processes is crucial for several reasons. Person to person interactions foster empathy and trust, which are essential for a positive workplace culture. AI can streamline processes, but it cannot replace the personal touch that builds strong relationships between employees and managers. People are also a backstop against potential bias resulting from AI. Human oversight is essential to ensure that people and talent decisions are fair and inclusive. Additionally, some HR decisions require a level of nuanced understanding and judgment that AI solutions can’t (yet) fully comprehend. Human involvement and leadership are still needed for complex interactions like conflict resolution, coaching, and performance assessment. 

For example, an AI-powered hiring system may flag a candidate as a low match based on algorithmic assessments, but a hiring manager should have the ability to override this decision after considering the candidate’s unique experiences and potential. Similarly, AI-driven performance evaluations should be cross-checked with direct feedback from managers and peers to ensure that qualitative factors like leadership potential or contributions to team culture are not overlooked. Organizations that rely too heavily on AI risk dehumanizing their HR processes, which can lead to unintended exclusionary practices. 

3. Diversify Data Inputs and Refine AI Models

AI systems are only as good as the data they are trained on. If the data reflects past inequities, the AI will reinforce those patterns rather than correct them. To counteract this, organizations must diversify their training data by incorporating inputs from a wide range of employee backgrounds, industries, and experiences. This includes ensuring that datasets represent different genders, ethnicities, educational backgrounds, and career paths. Some strategies organizations and HR technology providers can adopt to diversify training data include collecting data from a wide range of demographic characteristics, conducting regular audits on training data to check for gaps in data diversity, and adopting more inclusive approaches to collecting data to prevent homogeneity in the data sets. 

Beyond diversifying data, companies must also commit to ongoing training and refinement of AI models. AI systems should not be static; they need to be updated regularly to adapt to shifting workforce demographics and evolving DEI priorities. A best practice is to implement continuous monitoring mechanisms that flag disparities in AI-generated recommendations over time. For example, if an AI-driven internal mobility tool consistently suggests leadership roles to employees from one demographic group while overlooking others, the algorithm should be retrained to account for potential biases in its decision-making. 

Create AI Governance and DEI Oversight Committees 

HR leaders should not bear the responsibility of AI oversight alone. Instead, organizations should establish AI governance committees that include cross-functional representation from HR, IT, DEI leaders, legal teams, and data scientists. These committees should be tasked with developing ethical AI policies, conducting regular audits, verifying technology suppliers’ compliance with their AI tools, and ensuring compliance with emerging AI regulations. 

A strong governance framework ensures that AI decisions are not made in isolation but are aligned with broader organizational goals for DEI and fairness. For instance, a committee might establish guidelines requiring vendors to provide explainable AI models – meaning HR teams should understand why an AI tool made a particular recommendation, rather than accepting it as a black box decision. Additionally, DEI oversight teams can develop escalation procedures that allow employees to challenge AI-driven decisions, ensuring that individuals have a clear path to recourse if they believe an AI tool has made an unfair assessment. 

Educate HR Teams & Leaders on AI Literacy 

One of the biggest challenges in AI adoption is that many HR professionals lack the technical knowledge to critically assess AI tools. Without proper AI literacy, HR teams may unknowingly deploy biased algorithms or fail to ask the right questions when evaluating AI vendors. Organizations must invest in training programs that equip HR leaders with the knowledge needed to understand AI’s capabilities, limitations, and ethical implications. 

AI literacy training should include guidance on how AI models work, what common biases exist, and how to interpret AI-driven insights. Additionally, HR professionals should be trained to recognize when human judgment should take precedence over algorithmic recommendations. Some companies are beginning to integrate AI ethics training into their HR certification programs, ensuring that AI literacy becomes a core competency for HR leaders. When HR teams are empowered with AI knowledge, they can act as informed advocates for fairness and inclusion rather than passive users of technology. 

Sources of information for HR professionals on AI technology are increasingly available from HR professional organizations, from HCM technology providers, from independent analyst reports, and from open-source educational platforms like EdX and Coursera. Internally, HR can organize interactive workshops, targeted training programs for HR business areas, and collaborate with legal, IT, and operations teams for regular updates on AI developments and any changes in organizational policies. 

Conclusion: The Future of AI in HR – A Tool for Inclusion or Exclusion? 

AI has the power to be a force for good in HR, but only if companies take proactive steps to manage bias. AI presents both opportunities and risks for DEI in HR. If left unchecked, AI can amplify existing biases and create new barriers to workplace inclusion. However, when implemented with care, transparency, and human oversight, AI has the potential to drive more equitable decision-making and uncover insights that improve workforce diversity and inclusion.