The power of AI, such as ChatGPT, lies in its ability to parse queries and provide information based on vast amounts of text-based data. As the publisher of the upcoming book, Black Women’s Pathways to Executive Academic Leadership, by Crystal R. Chambers, Ph.D., we set out to explore how AI would respond to how Black women could accelerate their path to leadership in academia based on its algorithms and access to millions of documents, articles, and books.

We then asked the same question of White women and again of men in general. The results of the query brought to light significant disparities in the advice given to Black women compared to their White counterparts, exposing the underlying assumptions and biases embedded in the way we describe and evaluate people of color.

Unpacking the Assumptions

Upon analyzing the results below, it became evident that the advice provided to Black women emphasized a multitude of actions aimed at seeking more qualifications and skills, offering to do more service-oriented activities, and cultivating perseverance and resilience. Steps such as seeking continuous professional development, finding opportunities for ‘stretch’ assignments, and addressing biases and stereotypes highlighted the need for Black women to navigate a complex web of challenges to improve their chances of promotion.

On the other hand, the guidance given to White women and men appeared comparatively straightforward and simplistic, encouraging them to focus on performing their job well, doing research, and even maintaining a healthy work-life balance. See the full list of suggestions given by ChatGPT, by group, and in order below.

Our ChatGPT Query:

What are the best pathways for [Black women, White women, men] to get promotions in academia?

The responses:

The suggestions for White men, exactly mirrored those for White women, except for seeking family-friendly policies in seeking work/life balance (absent from a Black woman’s to-do list). This discrepancy in the advice raises concerns about the assumptions surrounding the capabilities and worth of Black women compared to White individuals.

The Reality

The truth is Black women have a long history of involvement in education. From the era of slavery, where they risked their lives to educate others, to the civil rights movement, where they fought for equal access to quality education, Black women have been at the forefront of educational activism.

• Representation and Role Models: Black women in educational leadership serve as important role models for students, particularly those from underrepresented backgrounds. Seeing someone who looks like them in positions of power and authority can inspire young Black girls to pursue leadership roles in education and beyond.
• Breaking Barriers: Black women in educational leadership have shattered glass ceilings and defied societal expectations. They have overcome racial and gender biases to rise to positions such as principals, superintendents, college presidents, and educational policymakers.
• Educational Advocacy: Black women in educational leadership have been influential advocates for educational equity and social justice. They work tirelessly to address disparities in educational resources, opportunities, and outcomes for marginalized students, particularly those from racial and ethnic minority groups.
• Culturally Responsive Leadership: Black women leaders bring a unique perspective and understanding of the diverse needs of students. They prioritize culturally responsive practices, incorporating the experiences, cultures, and histories of students into the educational framework to create inclusive and empowering learning environments.
• Mentorship and Support: Black women leaders often serve as mentors and provide support to aspiring educators and leaders. They recognize the importance of nurturing and guiding the next generation of Black women in educational leadership, ensuring continuity and growth in representation.
• Research and Scholarship: Black women in educational leadership contribute to research and scholarship that addresses the specific challenges faced by marginalized students. Their work examines the intersectionality of race, gender, and educational experiences, generating knowledge and insights that inform policy and practice.
• Collaboration and Networking: Black women in educational leadership actively engage in professional networks, organizations, and associations dedicated to advancing diversity, equity, and inclusion in education. They collaborate with colleagues to share best practices, exchange ideas, and collectively work towards transformative change.
• Resilience and Empowerment: Black women leaders in education demonstrate remarkable resilience in the face of adversity. They draw upon their experiences, strength, and determination to create positive change and empower others, ensuring that the voices of marginalized communities are heard and valued.
• Continued Advocacy: Black women in educational leadership continue to advocate for change at various levels, including policy reform, curriculum development, and community engagement. They strive to dismantle systemic barriers and create inclusive educational systems that provide equitable opportunities for all students.

The existence of such disparities in AI-generated content can be traced back to the data on which the AI model was trained. The training data for ChatGPT, like many other AI models, is sourced from a vast array of unstructured text available on the internet, including websites, books, and other sources. Since the AI model learns from patterns present in the data, any biases present in the training data are likely to be reflected in the model’s responses.

Thus, the biases found in the results may reflect the wider societal biases and stereotypes that have historically marginalized people of color. These biases can perpetuate harmful assumptions and further entrench systemic inequalities.

Is AI a Reflection of Our Society?

The question arises, does AI truly show us who we are as a society? The answer is both yes and no. AI can reflect the existing biases and prejudices present in society because it learns from the data it is exposed to. However, it is essential to recognize that AI is not sentient and does not possess personal beliefs or intentions. Instead, it reflects the data it was trained on, including both positive and negative aspects of human knowledge and behavior.

In this case, the AI’s response reveals how deeply rooted gender and racial biases can be in our collective knowledge. It is a stark reminder of the disparities faced by marginalized communities, especially Black women, who are often subject to multiple layers of bias.

But more importantly, we have to acknowledge that the overwhelming body of knowledge that has been written, published, widely distributed―and now, the source of AI processing and learning will continue to perpetuate stereotypes unless we diversify and expand our sources of information to include books, texts, and articles written by people of color—particularly in the academy, where scholars of color remain in service of knowledge while being denied in production of (anti-colonial) knowledge.

As AI technologies continue to advance, it becomes crucial to be mindful of the potential biases and ethical implications they carry. While AI can be a powerful tool for information retrieval and analysis, it must be used responsibly and ethically.

Conclusion

The findings from the query conducted with ChatGPT have shed light on the assumptions and biases ingrained in our collective understanding of people of color, particularly Black women. As we leverage the power of AI, some will point out that it is our responsibility as individuals and as a society to critically examine the technology’s outputs, question underlying assumptions, and take proactive steps to mitigate biases. By doing so, we can foster a more inclusive and equitable AI-driven world where the potential for harm is better assessed and realized.

And, by the way, this article was also submitted to and written with the assistance of ChatGPT. Maybe it will learn better than we have.