B y 2030, artificial intelligence (AI) is expected to contribute more than $15 trillion to the global economy and jobs that involve machine learning — a critical subset of AI — are projected to grow 76 percent over the next 10 years. Yet, the diversity of the AI workforce is struggling to keep pace. To make AI equitable and safe for generations to come, we need to focus on diversif- ying its talent pool — now. Statistical reports show how homogeneous the current AI talent base is. A study from the consulting services company McKinsey shows that the average share of employees developing AI solutions at respon- dents’ organizations who identify as women is 27 percent, and those who identify as racial or ethnic minori- ties is 25 percent. Further, insight from a study published in Gender, Technology and Development found that “structural and gender imbal- ances in the AI workforce and the gender divide in digital and STEM skills have direct implications for the design and implementation of AI applications.” The inequities in STEM education, particularly for young people of color, women, and nonbinary individuals, are well-recognized. Deep-rooted societal biases have historically excluded those who don’t fit the mold of cisgender, white, and male, both in school science labs and in profes- sional STEM roles. I became involved in educational program development to enhance opportunities for individuals histor- ically marginalized from technolog- ical domains. Today at AI4ALL — a U.S. nonprofit dedicated to creating a more inclusive, human-centered AI discipline through education, men- torship, and training for historically excluded talent — my focus is on nurturing the upcoming generation
AI4ALL DEMOGRAPHICS 2022-23 537 Total student participants 49% Identified as women or nonbinary 44% Identified as Black, Latinx, or Native American/Indigenous being used in a capacity that furthers racism and gender discrimination. Creating an AI workforce represen- tative of our greater population is a necessary way to combat this issue and can only be achieved by widening from leading experts and policy- makers underscore the importance of steering AI development toward ethical, inclusive paths. For example, algorithmic bias is a growing dan- ger. Automated decision-making can amplify or exacerbate societal biases that exist in data. Such tools are of bright minds who will influence the trajectory of artificial intelligence (AI) and its applications. While the spotlight often shines on a few prominent figures like Elon Musk and Mark Zuckerberg as the faces of AI, the reality is that AI prog- ress is a collaborative effort involving diverse talents. Unfortunately, con- tributions from individuals outside of a narrow demographic often aren’t highlighted, leading to a skewed narrative and potentially broader ramifications: Many promising young people do not see themselves repre- sented among those who power AI. Nowhere in STEM is diversity more urgent than in AI. Recent warnings
SPRING 2024 • 41
Made with FlippingBook Digital Proposal Creator