Published on 8 March 2021, International Women’s Day, this policy briefing from the Turing’s Women in Data Science and AI project presents new research into gender gaps in AI and data science, and the extent and impact of men’s dominance in these fields.
What does the report reveal?
A decade ago, Harvard Business Review named data scientists as “the sexiest job of the 21st century.” Since then hiring in the fields of artificial intelligence and data science has exploded as the world is increasingly being built around smart machines and automated systems.
Yet the people whose work underpins that vision are far from representative of the society those systems are meant to serve.
Only 22% of data and AI professionals in the UK are women, and this drops to a mere 8% of researchers who contribute to the pre-eminent machine learning conferences.
This is not only a fundamental issue of economic equality, but also about how the world is designed and for whom. Mounting evidence (e.g. Buolamwini and Gebru, 2018; West, Whittaker and Chew, 2019) suggests that the under-representation of women and marginalised groups in AI results in a feedback loop whereby bias gets built into and amplified by machine learning systems.
Addressing the gender job gap in AI is the first step to ensuring that our technology works for all of society.
Existing data is sparse: Raw, intersectional industry data about gender diversity in the AI workforce is severely limited. The available high-level statistics, however, show there are fewer women working in the data and AI fields in the UK compared to the global average. Currently, women make up an estimated 26% of workers in data and AI roles globally, which drops to only 22% in the UK. Further, in the UK, the share of women in engineering and cloud computing is a mere 14% and 9% respectively.
Diverging career trajectories: There is evidence of persistent structural inequality in the data science and AI fields. Women are more likely than men to occupy a job associated with less status and pay in the data and AI talent pool, usually within analytics, data preparation and exploration, rather than the more prestigious jobs in engineering and machine learning.
Job attrition rates: Women working in AI and data science in the tech sector have higher attrition rates (i.e. leaving the industry) than men.
Self-reported skills: Men routinely self-report having more skills than women on LinkedIn. This is consistent across all industries and countries in the sample in the report. This correlates with existing research into women’s lower confidence levels in their own technical abilities.
The qualification gap: Women in data science and AI have higher formal educational levels than men across all industries. The achievement gap is even higher for those in more senior ranks (i.e. for C-suite roles).
Participation in online platforms: Women comprise only about 17% of participants across the online global data science platforms Data Science Central (‘DS Central’), Kaggle and OpenML. On StackOverflow, women are a mere 8%. Additionally, only about 20% of UK data and AI researchers on Google Scholar are women. Of the 45 researchers with more than 10,000 citations, only five were women.
The report shows the extent of gender disparities in careers, education, jobs, seniority, status and skills in the AI and data science fields. Utilising a new, curated dataset and an innovative methodology, it explores the gendered dynamics of careers in the UK and other countries in fresh detail.
These original findings add urgency to the drive to improve women’s opportunities in the technology industry and highlights the need for effective policy responses if society as a whole is to reap the benefits of technological advances.
This work has added urgency since the drive to close the gender gap in the technology industry risks being derailed by the pandemic.
What does the report recommend?
- The world's tech companies must improve their level of reporting regarding diversity and inclusion. Many companies currently only provide headline statistics. Currently most large technology firms are unwilling to disclose their own detailed diversity data. The lack of transparency has serious implications for Government policymaking around technological advancement and equity, and for labour market policies.
- The UK government should require tech companies to scrutinise and disclose the gender composition of their technical, design, management and applied research teams. This must also include mandating responsible gender-sensitive design and implementation of data science research and machine learning. This is an issue of social and economic justice, as well as one of AI ethics and fairness.
- Countries need to take proactive steps to ensure the inclusion of women and marginalised groups in the design and development of machine learning and AI technologies.
- Companies should implement gender inclusive labour market policies, such as paid maternity and parental leave and flexible working hours and affordable childcare must be provided. These measures are a prerequisite to ensuring that women’s disproportionate responsibility for domestic and care work does not inhibit their ability to participate in the digital economy on an equal footing to men.
- Companies in the tech sector must embed intersectional gender mainstreaming in human resources policy so that women and men are given equal access to well-paid jobs and careers. Actionable incentives, targets and quotas for recruiting, up-skilling, re-training, retaining and promoting women at work should be established, as well as ensuring women’s equal participation in ‘frontier’ technical and leadership roles.
As Professor Judy Wajcman, head of the project added: “Diversity in the fields of data science and AI is essential if we do not want to replicate social biases in technical systems.”
Read the full report: Where are the women? Mapping the gender job gap in AI by Erin Young, Postdoctoral Research Fellow in the Public Policy Programme at The Alan Turing Institute; Judy Wajcman, Principal Investigator of the Women in Data Science and AI project at The Alan Turing Institute, and Anthony Giddens Professor of Sociology at the London School of Economics; and Laila Sprejer, a Data Science Research Assistant in the Public Policy Programme at The Alan Turing Institute.
Cover photo credit: Mike Ngo Photography via Flickr