How AI Can Affect Bias in Leadership

AI Reflects and Amplifies Human Biases

One of the biggest revelations from the experiments was that AI models are only as unbiased as the data they are trained on. Nino, an Italian ALJ Guide, noted that different AI models—such as ChatGPT, Gemini, and Copilot—demonstrated biases when tasked with evaluating a complex business decision involving multinational teams.

The AI avoided making a final decision and recommended that a report be created for human decision-makers. Nino concluded that AI can be a useful tool for structuring decision-making but cannot independently resolve issues influenced by cultural, social, or financial complexities.

AI Can Be Trained to Recognize Cultural Differences, But It Has Limits

Vermont-based ALJ Guide Katharine’s experiment simulated leadership interactions across different cultural contexts. Initially, AI-generated personas lacked cultural nuance—defaulting to generic professional responses rather than reflecting the diverse perspectives of leaders from different regions. However, when trained with specific prompts about cultural differences, AI became more effective at predicting how leaders from various backgrounds might react to decisions.

The outcome? AI has the potential to enhance leadership self-awareness but it can also run the risk of reinforcing stereotypes if not properly trained—especially if you don’t explicitly prompt it to challenge you on that.

AI as an Augmenting Tool, Not a Decision-Maker

Sonny’s experiment wondered whether AI could understand the challenges of underrepresented leaders in the workplace. He found that AI, much like our human colleagues, often fails to recognize the depth of emotional and cultural barriers faced by leaders from minority backgrounds. This led to the realization that AI can offer broader perspectives and surface potential biases, but it still lacks the emotional intelligence and lived experience necessary to fully grasp the nuances of leadership.

How We Can Proceed—Ethically

Nino, Sonny, and Katharine’s experiments confirm that AI mirrors the biases of its creators and the data it is fed. If leaders are using AI to guide their decision-making, they must remember to embrace critical thinking of the output. Leaders cannot expect the latest AI bot to solve any bias that may exist; instead, they should use it as a tool to enhance their awareness, challenge their assumptions, and make more inclusive decisions.

Attend a Future ALJ Global Masterclass!

You’re invited if you’ve ever participated in any ALJ workshop or program! Sign up for the alumni mailing list to receive calendar invites for future events, newsletters, and other resources to guide your leadership journey.