Editorial
Artificial intelligence (AI) and large language models (LLMs) are transforming the world in unprecedented ways. From health care to education, from business to entertainment, from politics to social justice, AI and LLMs are reshaping every sector of human activity. However, these powerful technologies also pose significant ethical risks and challenges, such as bias, discrimination, privacy, accountability and human dignity. How can we ensure that AI and LLMs are aligned with the values and interests of humanity?
Here, we will examine the current state of women's representation and participation in the AI and LLM fields, the barriers and challenges they face and the opportunities and benefits they bring. We will also explore recommendations and best practices for promoting gender diversity and inclusion in the AI and LLM ecosystem.
Challenges and Issues of Bias, Transparency and Ethics in the Learning Process of LLMs
AI systems that can produce natural language texts based on some input, such as a word, phrase or sentence, are LLMs. LLMs learn from huge amounts of text data from different sources, such as books, news articles, social media posts and web pages. LLMs can do various tasks, such as answering questions, summarizing texts, writing essays, translating languages and making chatbot conversations. LLMs have the possibility to improve human communication, creativity and knowledge, but they also bring significant challenges and risks for society.
One of the prevailing issues in AI development is the problem of bias. Bias refers to the systematic deviation of an AI system's output from the expected or desired outcome. Bias can result from various factors, such as the quality, quantity and diversity of the data used to train the AI system, the design and implementation of the algorithms and models and the interpretation and evaluation of the results. Bias can have negative impacts on individuals and groups, such as discrimination, exclusion, stereotyping and misinformation. For example, LLMs can generate texts that reflect or amplify the biases present in their training data, such as gender, racial or cultural stereotypes, or harmful or offensive language.
Another issue in AI development is the lack of transparency. Transparency refers to the extent to which an AI system's inputs, outputs, processes and decisions can be understood and explained by humans. Transparency is important for ensuring the reliability, validity and accountability of AI systems as well as for building trust and confidence among users and stakeholders. However, many AI systems, especially LLMs, are often opaque and complex, making it difficult to trace and justify how they generate their outputs or to detect and correct any errors or anomalies. For example, LLMs can produce texts that are fluent and coherent but also inaccurate, misleading or irrelevant, without providing any indication or evidence of their sources or methods.
A third issue in AI development is the ethical dilemma. An ethical dilemma refers to the situation where an AI system's output or action involves a conflict or trade-off between different moral values, principles or interests. Ethical dilemmas can arise from the uncertainty, complexity and ambiguity of the real-world contexts and scenarios where AI systems are deployed and used. Ethical dilemmas can also stem from the diversity and plurality of the perspectives and expectations of the users and stakeholders of AI systems, such as developers, regulators, consumers and society at large. For example, LLMs can generate texts that are useful and beneficial for some purposes or groups but also harmful or detrimental for others or that violate some ethical norms or standards, such as privacy, fairness or human dignity.
See more: Dealing With AI Biases, Part 1: Acknowledging the Bias
The Critical Need for Diverse Perspectives
The issues of bias, transparency and ethical dilemma in AI development are not merely technical or scientific problems, but also social and cultural ones. They reflect the values, assumptions and interests of the human agents who create, use and govern AI systems. Therefore, it is essential to have diverse perspectives and voices in the AI and LLM fields, especially those of women, who constitute half of the world's population and experience unique challenges and opportunities in the digital age.
Women bring invaluable perspectives to AI development, as they can offer different insights, experiences and expertise that can enrich and improve the quality, relevance and impact of AI systems. Women can also contribute to the ethical and responsible design and use of AI systems, as they can raise awareness, advocate and address the social and moral implications and consequences of AI outputs and actions. Moreover, women can foster a culture of diversity and inclusion in the AI and LLM ecosystem, as they can create and support networks, communities and initiatives that empower and mentor other women and underrepresented groups to participate and thrive in the AI and LLM fields.
However, despite the critical need and benefit of women's perspectives in AI development, women are still underrepresented and marginalized in the AI and LLM fields. Only 22% of AI professionals globally are women, and only 12% of AI researchers are women, according to a recent report by the World Economic Forum. Moreover, women face various barriers and challenges in entering and advancing in the AI and LLM fields, such as gender stereotypes, discrimination, harassment, bias, lack of role models, mentors and sponsors and unequal access to education, training, resources and opportunities.
A computer scientist and activist who founded the Algorithmic Justice League, an organization that strives to uncover and reduce the harms of biased and discriminatory AI systems, is Joy Buolamwini, one of the women leaders who have made significant contributions to ethical AI. Buolamwini's work on facial recognition biases shows that many commercial facial analysis tools are more likely to make mistakes for faces that are darker-skinned or female than for faces that are lighter-skinned or male, causing serious problems of fairness, accuracy and accountability. Buolamwini's research and advocacy have raised public awareness and policy reactions to address the social and ethical impacts of facial recognition technologies, such as the bans or moratoriums on the use of facial recognition by law enforcement agencies in several cities and states in the U.S.
Timnit Gebru is an example of women in AI who have advocated for inclusivity, transparency and ethical considerations. She is a former co-leader of Google's Ethical AI team and a co-founder of Black in AI, a community that supports the representation and participation of Black people in AI. Gebru's research and activism have focused on the racial and gender diversity of AI researchers and practitioners as well as the environmental and societal costs of large-scale AI models and data sets. Gebru's work has shown the need for more scrutiny and accountability of the power and influence of big tech companies over the development and deployment of AI systems as well as the importance of involving and empowering marginalized and affected communities in the design and governance of AI systems.
These are just two of the many examples of women who are at the forefront of ethical AI, who have shown courage, leadership and innovation in advancing the values and principles of human-centered and responsible AI. Their perspectives and voices are essential for ensuring that AI systems are not only intelligent and efficient, but also fair, inclusive, transparent and beneficial for all.
See more: Dealing With AI Biases, Part 2: Inherited Biases From Data
Challenges Women Face in the Tech Industry
A major obstacle for women in the tech industry is the low representation and diversity of science, technology, engineering and mathematics (STEM) fields, especially in AI. Only 35% of STEM students in higher education worldwide are women, and only 3% of female students choose to study information and communication technology (ICT), according to a 2018 report by UNESCO. Furthermore, women make up only 28% of researchers in STEM fields globally, and only 25% of leaders in the tech sector. These numbers indicate that women are greatly underrepresented and undervalued in the STEM and AI workforces, which affects their chances, impact and recognition in the fields.
Another challenge women face in the tech industry is the gender gap in technology and the consequences of underrepresentation. The gender gap in technology refers to the disparity between men and women in terms of access, use, creation and impact of ICT and AI. This gap can have negative effects on women's economic, social and political empowerment as well as their participation and contribution to the digital society. For instance, the gender gap in technology can exacerbate existing inequalities and biases in AI systems, such as the ones revealed by Buolamwini and Gebru, which can affect women's rights, safety, privacy and dignity. Furthermore, the gender gap in technology can hinder women's potential and innovation in AI development, as they are less likely to have the skills, resources, networks and support needed to pursue and excel in the field.
Therefore, it is imperative to address the challenges women face in the tech industry and promote their inclusion and empowerment in the AI and LLM fields. Women have much to offer and gain from the advancement of AI, as they can bring diverse perspectives, experiences and expertise that can enhance and improve the quality, relevance and impact of AI systems. Women can also play a vital role in ensuring that AI systems are ethical and responsible, as they can advocate and address the social and moral implications and consequences of AI outputs and actions. Moreover, women can foster a culture of diversity and inclusion in the AI and LLM ecosystem, as they can create and support networks, communities and initiatives that empower and mentor other women and underrepresented groups to participate and thrive in the AI and LLM fields.
See more: Dealing With AI Biases, Part 3: Emergent Biases in Operational Models
The Path Forward
Mentorship Programs
Mentorship programs can provide women with guidance, advice, support and inspiration from other women and allies who have experience and expertise in the AI and LLM fields. Mentorship programs can also help women develop their skills, confidence and network as well as access opportunities and resources that can advance their careers and goals in the fields. For example, Women in AI and Women in Machine Learning are two global communities that offer mentorship programs for women in AI as well as events, workshops and publications that showcase and celebrate women's achievements and contributions in the field.
Policy Changes
Policy changes can address the structural and systemic barriers and challenges that women face in the tech industry, such as the gender pay gap, lack of parental leave and childcare support, harassment and discrimination in the workplace and underrepresentation in leadership and decision-making positions. Policy changes can also create incentives and regulations that encourage and ensure the inclusion and diversity of women and other marginalized groups in the AI and LLM workforces as well as the ethical and responsible development and deployment of AI systems. For example, the European Union's Gender Equality Strategy 2020-2025 aims to close the gender gaps in digital and STEM fields as well as promote gender mainstreaming and impact assessment in AI policies and initiatives.
Corporate Commitments to Diversity
Corporate commitments to diversity can demonstrate and implement the values and principles of inclusion and diversity in the tech industry as well as to foster a culture of respect, trust and collaboration among employees and stakeholders. Corporate commitments to diversity can also involve setting and monitoring targets and indicators of diversity and inclusion as well as providing training, awareness and accountability mechanisms that ensure the implementation and evaluation of these commitments. For example, Google's Diversity Annual Report 2020 outlines the company's progress and challenges in advancing diversity, equity and inclusion in its workforce, products and community as well as its actions and initiatives to address them.
Education Programs
Education programs can encourage and equip young women and girls to follow their interests in technology as well as to give them the knowledge, skills and opportunities that can get them ready for the future of work and society. Education programs can also address and change the stereotypes and biases that prevent and exclude women and girls from the STEM and AI fields as well as demonstrate and highlight the role models and mentors that can inspire and support them. For instance, Girls Who Code is a nonprofit that works to close the gender gap in technology by teaching girls coding and computer science skills as well as building a community of female tech leaders and innovators.
Webinar
Sep
25
CMS Briefing: A Live Look at What’s Next in AI-Driven Platforms
Learn how leading organizations are using AI‑driven tools to publish faster, personalize smarter and stay secure.
Register
Webinar
On demand
Ready or Not: How Data-First Organizations Are Unlocking Agentforce Potential
Learn how to cut through the noise, activate Agentforce and build a Salesforce AI strategy that actually delivers.
Watch Now
Webinar
On demand
AI in Customer Service: Faster Resolutions, Happier Customers
Don’t let rising demand burn out your team. See how to build a smarter, more resilient support org.
Watch Now
Webinar
On demand
From Hype to High-Impact CX Strategies That Actually Scale
Turn buzzworthy AI and outsourcing trends into measurable CX wins with fresh 2025 data.
Watch Now
Webinar
On demand
Insights to Action Rethinking the Contact Center for Real Business Impact
Join our exclusive webinar to hear CX executives share their innovative strategies for transforming service delivery.
Watch Now
Webinar
On demand
Accelerating Healthcare Ops with AI
Sign up to see how your peers are putting AI to work — and making it stick.
Watch Now
Call to Action
It is important for women to lead in AI and LLM, not only because it is fair and equal, but also because it is innovative and excellent. Women have the right and the ability to influence the future of AI and to enjoy its benefits and face its challenges. However, women need more help and acknowledgement from the tech industry, policymakers, educators and society in general to overcome the obstacles and prejudices that limit their involvement and progress in the field. Everyone can contribute to supporting and promoting women's leadership in AI and LLM, by acting in your own areas of impact and accountability. Whether you are a researcher, developer, manager, teacher, student or citizen, you can make a change by:
- Raising awareness and educating yourself and others about the importance and impact of women's leadership in AI and LLM as well as the challenges and opportunities they face in the field.
- Mentoring and sponsoring women and girls who aspire to pursue careers in AI and LLM as well as providing them with the guidance, feedback, resources and opportunities that can help them achieve their goals and aspirations.
- Supporting and joining the communities and networks that promote and celebrate women's achievements and contributions in AI and LLM as well as providing them with the space, voice and visibility that they deserve.
- Advocating and influencing the policies and practices that foster and ensure the inclusion and diversity of women and other marginalized groups in the AI and LLM workforces as well as the ethical and responsible development and deployment of AI systems.
By supporting and advocating for women's leadership in AI and LLM, we are not only empowering women, but also enhancing and improving the quality, relevance and impact of AI systems. We are also contributing to a fair and equitable future, where AI serves the needs and values of all people and where everyone can benefit from its potential and opportunities. Women's leadership in AI and LLM is not a choice, but a necessity. Let's make it happen together.
See more: Dealing With AI Biases, Part 4: Fixing the Root Cause of AI Biases
Learn how you can join our contributor community.
