Practical Strategies for Leaders, HR Professionals, and People Managers
Artificial intelligence (AI) tools such as Microsoft Copilot, ChatGPT, Claude, and Gemini are quickly becoming part of everyday work. Leaders are using these generative AI tools to draft communications, develop training materials, write job descriptions, summarize meetings, and support decision-making.
While AI can increase efficiency and save time, it is important to remember that these tools learn from data created by people. When that data contains historical biases, stereotypes, or gaps in representation, AI can reflect or reinforce those same patterns. Research has found that generative AI systems can reproduce biases related to gender, race, and other aspects of identity if users are not intentional about their review process.
Reducing AI bias is not simply a technology issue. It is an inclusive leadership issue. At Archbright, we talk often about creating workplaces where employees feel respected, valued, and included. As AI becomes more integrated into workplace practices, those same principles should guide how we use technology.
Organizations that apply an inclusive lens to AI are more likely to create communications, policies, and employee experiences that reflect their values while minimizing unintended harm.
Inclusive and responsible use requires leaders to treat AI as a drafting partner, rather than a final decisionmaker. AI can help generate ideas, organize information, and create first drafts. However, experts consistently emphasize that decisions involving hiring, promotions, accommodations, performance management, or employee relations should always involve human review and judgment.
The quality of an AI response depends heavily on the instructions it receives.
Instead of asking AI to "write a leadership profile," try:
"Write a leadership profile using inclusive language, avoiding stereotypes, and highlighting diverse leadership styles."
You can also ask AI to use accessible language, incorporate diverse examples, and consider multiple perspectives.
Many users stop after the first response. Instead, use follow-up prompts such as:
"Review this content for assumptions, stereotypes, exclusionary language, or missing perspectives."
Research suggests that prompting AI to think about bias can help improve the inclusiveness of AI-generated content.
AI often defaults to the most common patterns in its training data. When generating workplace scenarios, request examples that include people from different generations, backgrounds, abilities, cultures, and leadership styles.
This helps create content that better reflects today's workforce and reduces the risk of unintentionally excluding employees.
AI-generated content may sound polished, but that doesn't mean it's free from bias.
Review content for:
Gendered assumptions
Ableist language
Unnecessary references to protected characteristics, such as age
Cultural stereotypes
Exclusionary examples such as asking AI to create a company party invite and its output is, “All employees must wear a suit or a dress” for attire.
Applying the same communication standards you expect from employees to AI can improve the quality and inclusiveness of its output.
One of the most effective ways to improve AI outputs is to provide better context.
Many organizations can now create custom AI agents that reference approved organizational resources. Rather than relying solely on publicly available information, these tools can be grounded in organizational values and guidance.
Consider providing:
Providing trusted examples can help AI generate content that better aligns with your organization's culture and expectations.
Organizations don't need a complex AI strategy to make progress.
Start by:
Small changes in how people use AI can significantly impact the quality and inclusiveness of the content it produces.
Like any tool, generative AI reflects the information and assumptions on which it is built. As leaders, we must work to ensure the technology we use reflects the values we want our organizations to uphold.
By applying an inclusive lens to AI, leaders can help create communications, policies, learning materials, and workplace experiences that are not only more effective but also more inclusive. AI may generate the first draft, but inclusive leadership should always have the final word.
Author's Note: This article was developed in anticipation of Archbright's October IDEAL webinar on Reducing AI Bias and Creating More Inclusive Workplace Content. The webinar will provide additional practical strategies for leaders, HR professionals, and people managers looking to use AI responsibly while supporting a more inclusive workplace. Archbright members can attend for free, email info@archbright.com if you need a code.
Sources:
Prompt coaching tool raises user awareness of bias in generative AI systems | Penn State University
About the Author:
Jenna Shellman is a seasoned corporate trainer and a diversity, equity, and inclusion (DEI) consultant. As the IDEAL (Inclusion, Diversity, Equity, Accessibility, and Leadership) Director at Archbright, Jenna partners with organizations to embed DEI into their leadership practices and workplace culture. With a dynamic approach to training, she designs and facilitates impactful workshops that promote professional growth, inclusive leadership, and psychologically safe work environments. Jenna’s work empowers both individuals and organizations to embrace transformation, build stronger teams, and foster workplaces where all employees can thrive. She leads with a deep commitment to inclusion and a vision for long-lasting cultural change.