Artificial Intelligence Biases Being Instilled Through Lack of Tech Diversity
In a world where Artificial Intelligence (AI) is becoming increasingly prevalent, concerns about the lack of diversity in the AI workforce and the potential for AI bias have come to the forefront. A recent report by Georgetown University reveals that Blacks and Hispanics are underrepresented in the AI workforce, with men holding a staggering 80% of tenured faculty positions in university AI departments globally.
Prof. Safiya Noble, a professor at the University of California Los Angeles, expresses concern that the government's attack on Diversity, Equity, and Inclusion (DEI) initiatives will undermine efforts to create opportunities in AI for marginalized groups. This concern is further exacerbated by the backlash against movements like Black Lives Matter and allegations of anti-conservative bias.
However, there are global initiatives aimed at addressing these issues. One such initiative is the focus on education, training, and apprenticeships for underrepresented groups. Governments and organizations are creating AI-focused educational and apprenticeship programs to grow a diverse talent pool. For instance, the U.S. AI Action Plan includes incentives for AI education and training to broaden workforce participation.
Another strategy is the development of ethical guidelines and bias mitigation in AI development. There is a growing emphasis worldwide on identifying and reducing biases embedded in AI models, often caused by unrepresentative training data or flawed assumptions. Various international coalitions and bodies work on standards and best practices to ensure fairness and accountability in AI.
Inclusive hiring and retention initiatives are also being promoted by tech companies and governments. These include mentorship, bias training for recruiters, and transparency in career progression. Amazon.com, for example, scrapped an AI recruiting tool when it found it was selecting resumes favoring men over women.
Several governments integrate diversity goals into AI policies. While the Trump Administration’s 2025 AI Action Plan focuses primarily on innovation, infrastructure, and global competition without explicit emphasis on diversity initiatives, other nations and international organizations more explicitly target workforce diversification as part of broader AI governance goals.
International forums and alliances encourage cooperation on AI ethics, including diversity and bias mitigation in AI development and deployment. One such organisation is AI4ALL, a non-profit working to develop an inclusive pipeline of AI professionals. Since 2015, AI4ALL has helped 7,500 students.
Despite these efforts, the progress has been slow. From 2021 to 2024, the number of women working in AI globally increased by only 4%. Maya De Los Santos, an Afro-Latina woman with degrees in computer and electrical engineering, is one of those who are determined to change this. She is interested in a career in AI to ensure marginalized communities are protected from AI risks and understand its benefits.
The story, published with permission from Thomson Reuters Foundation, is relevant to topics such as education, jobs, inequality, gender equality, women, innovation, inclusion, artificial intelligence, and is relevant to the SDGs 4, 5, 8, 9, 10, 16, and 17. Among AI technical occupations, Hispanics hold about 9% of jobs, while Black workers hold about 8%. However, underrepresented groups such as Blacks, Hispanics, and Native Americans are less likely to have access to AI classes in public high schools.
In conclusion, while major AI policy frameworks like the recent U.S. AI Action Plan prioritize rapid technological innovation, deregulation, and global leadership with limited direct focus on diversity and bias reduction, the broader global landscape includes initiatives promoting inclusive education, ethical AI practices, inclusive hiring, and international cooperation to increase diversity in the AI workforce and mitigate bias in AI systems. However, more targeted diversity and bias mitigation efforts are primarily driven by non-governmental entities, academia, and international organizations beyond these broad national policies.
- Prof. Safiya Noble is concerned that government attacks on Diversity, Equity, and Inclusion (DEI) initiatives could hinder efforts to create opportunities in AI for marginalized groups.
- The U.S. AI Action Plan includes incentives for AI education and training to broaden workforce participation, aiming to grow a diverse talent pool.
- Amazon.com scrapped an AI recruiting tool when it found it was selecting resumes favoring men over women, demonstrating the importance of bias training.
- AI4ALL, a non-profit organization, works to develop an inclusive pipeline of AI professionals, focusing on education, diversity, and bias mitigation in AI development and deployment.
