Women in Artificial Intelligence – A Visual Study of Leadership Across Industries

Daniel Faggella

Daniel Faggella is Head of Research at Emerj. Called upon by the United Nations, World Bank, INTERPOL, and leading enterprises, Daniel is a globally sought-after expert on the competitive strategy implications of AI for business and government leaders.

women in artificial intelligence

This article was originally written in 2017 by Lauren D’Ambra, former editor at Emerj.com.

Women in artificial intelligence (AI) and machine learning (ML), or the lack thereof, is not a new topic in media, just as gender equity and disparity in the workplace is not a new subject of research for academics and think tanks. But discussing these issues openly is no less important. While we address the potential reasons and implications of these issues toward the end of this article, our initial interest in this subject came from our desire to know the following:

  1. How many women are in C-level and other leadership roles in the AI and ML industries compared to males?
  2. How might these numbers compare to other industries and the workforce at large?
  3. What are the potential implications of female presence (or lack thereof) in leadership roles within AI and ML companies?

We looked at top-level female presence (see definition below), and more specifically C-level roles, across 287 companies (average size between one and 100 employees) in seven core industries that leverage AI and ML technologies as part of their main product and service delivery. While a small segment of studies have been done in the past on female founders, including Crunchbase’s twice-published study on primarily U.S.-based companies, our study focuses on C-level roles, specifically within companies and industries applying ML and AI as a primary technology or solution.

The data set that we used to help answer the first question is based on the 2016 machine learning landscape published annually (since 2014) by Bloomberg Beta’s Shivon Zilis and James Cham, one that we follow regularly in our overview of the AI industry. We encourage readers to visit our methodology section for additional information on this landscape and for more details regarding our research approach.

While not a fully-inclusive study, we share our findings as a representative example of where women leaders are in AI and ML at present, and explore potential reasons for gaps and what a more equitable gender presence could mean for industry at large.

Note on terminology:

Top-level female presence: Defined as females in an executive-level role (i.e. chairpersons, directors, presidents/vice presidents), including C-level executives. Used when referencing percentage of companies within a particular industry or across industries. This term differs from the more specific data we collected on just C-level executive roles.

Additionally, because two industry names/categories chosen by Bloomberg Beta are similar at first glance—enterprise functions and enterprise intelligence—we provide a brief definition of each:

Enterprise functions: Service industry applications, such as customer support and marketing.

Enterprise intelligence: Specific ML-based technologies that serve as fundamental building blocks and drivers of applications, like computer vision and sensor technology.

Research Highlights at a Glance:

  1. Technology stack (as an industry) has a notable lack of females, both in top-level female presence and C-level roles, across sub-industries and the broader industry.
  2. Vision (technology stack) and imaging (healthcare) are both low in female leadership.
  3. The most “popular” C-level position amongst females, in terms of total number, is CEO; but in terms of ratios, females are the minority across all C-level roles—with the exception of human resources and talent.
  4. Our finding of 18% female C-levels across AI/ML companies is on par with other similar findings across industries.

In the next few sections, we break down in detail the results of our researching findings.

The Landscape of Women Leaders in AI/ML Industries

Based on our data set, the most populated industries in terms of top-level female presence are enterprise functions, autonomous systems, and industry at large (see chart below). Interestingly, two of these same industry categories have some of the biggest gaps in number of female C-level executives—autonomous systems and industry (which includes sub-industries like agriculture, education, etc.)—alongside agents and technology stack companies.

Industries with the biggest gaps in top-level female presence include enterprise intelligence and technology stack. Considering that those companies categorized as technology stack are lacking both overall female and C-level female presence with just 13% for both categories, it’s the clear tortoise in the pack. 

Chart 1.1


Note on Chart 1.1: All numbers represented in roll-over menu are percentages. Top-level female presence refers to percentage of companies with female leaders within each industry, while C-level female executives refers to percentage of female-held roles within each industry.

Healthcare, enterprise functions, and enterprise intelligence industries have the most female C-level leaders overall. In looking at most female C-level executives, we drill down to the sub-industry level. A larger number of women in C-level roles are found in patient-oriented health care application companies, recruiting and security companies (which fall under the enterprise functions industry), and internal data and audio companies (under enterprise intelligence).

Chart 1.2



Two sub-industries lacking female C-level executives are visual applications in enterprise intelligence and imaging in healthcare. The overlap in imaging technology applications seems potentially significant in its implications of lack of upper-level female involvement (see further discussion section). Ironically, both of these sub-industries fall under overarching industries that maintain a higher-level female presence overall.

We previously noted the low level (13%) of females holding C-level positions with technology stack companies, including the sub-industries natural language (5%), agent and conversational interfaces (8%), and development (13%). Notably, the legal sub-industry had a 0% female presence, the only sub-industry in which no females were found in either C-level or other leadership positions.

Chart 1.3

Note on Chart 1.3: Total number of women filling C-level roles across industries (see our methodology).

In terms of sheer number, there were more women CEOs—a total of 20—across industries than in any other C-level positions. Several of these CEO positions were filled by founders, though we didn’t track overlap (see further discussion section). This number drops by almost half for the next most populated C-level roles—CMO (marketing) and COO (operations).

When considering the 33 total C-level roles we documented in our research, it’s significant that many of these positions were not filled by any females or had a scarce female presence (i.e. one or two total) compared to a greater number (five or more) of male counterparts. Several C-level roles, such as chief of sales and chief of content, were “unique” in that there were filled by only one or two individuals across all industries; we did not give much notice to these instances, though more information can be found on such roles in our included methodology section. We did think it important to list all C-level roles (chart 1.3) identified in this study that were filled by at least one female.

Chart 1.4

[Draft] Women in AI 9
Role Role Description Total C-Level Executives No. Female No. Male
CEO Chief Executive Officer 156 20 136
CTO Chief Technology Officer 94 9 85
COO Chief Operating Officer 46 11 35
CFO Chief Financial Officer 35 10 25
CSO Chief Science Officer (Chief Scientist) 26 1 25
CMO Chief Marketing Officer 22 11 11
CCO Chief Commercial Officer 22 0 22
CPO Chief Product Officer 17 1 16
CRO Chief Revenue Officer 11 0 11
Chief People Officer, Chief Human Resources, Chief Talent Officer 10 6 4


We placed emphasis on roles that had a more significant presence of executives in total (10 or more) and the resulting ratio between female and males (chart 1.4). Perhaps not surprisingly, by sheer number alone CEO was the most popular role for females and males overall—a title that almost all companies include as part of their executive team—with 156 individual CEOs in total. When compared to the whole, females hold only 12.82% of the CEO positions across industries. Ultimately, there was no one C-level role in which females dominated over males. The largest proportion of females in C-level roles was 60% CHOs (human resources/staff/talent).

This data, considered on its own, points to stark gaps in female C-level presence in AI and ML companies across industries. To provide some context for our findings, we sought out additional studies on current averages of women executives and founders in tech and other industries. While the data is not apt for direct comparison, similar studies help illuminate potential patterns and similarities/differences in presence of female leadership across industries and various sizes of companies.

External Studies and Validation

According to a report done by the nonprofit research group Catalyst, as of 2017 about 5.8% of CEO roles in the S&P 500 are held by women, a rise of 1.8% since 2015. In the same 2017 report, about 25.1% of executive and senior-level management roles in the S&P 500 are filled by women. McKinsey’s “Women in the Workplace 2016” reports that the number of women C-levels in the corporate pipeline is 19%. Our data set yielded an average 18% females in C-level roles and an average 26% females in a top-level management and/or founding roles. While the industries and size of companies may differ, it’s interesting to note the almost on-par percentage of females in C-level roles in comparing our findings to those from McKinsey, a number that is winnowed significantly in the S&P 500. 

Emergent Questions

Based on our analysis of the data, the following key questions arose:

  • What are the factors that draw women toward and/or keep women from entering management-level and C-level positions?
  • What are the implications for companies, in the AI and ML space and beyond, with more or less female leaders?

To help us begin to address these questions, we again sought out reputable sources, including research groups and thought leaders who have addressed the topic, as well as female C-level executives themselves.

What are the factors that draw women toward and/or keep women from entering management-level and C-level positions?

In a survey based on 1,000 male/female respondents, Author Tara Mohr for the Harvard Business Review found that 78% of women’s reasons for not applying for a position have to do with “believing that the job qualifications are real requirements and seeing the hiring process as more by-the-book and true to the on paper guidelines than it really is.” Towards the close of her analysis, Mohr gives a memorable reflection of her own experience as a woman striving to find a place of leadership in enterprise:

“When I went into the work world as a young twenty-something, I was constantly surprised by how often, it seemed, the emperor had no clothes. Major decisions were made and resources were allocated based not on good data or thoughtful reflection, but based on who had built the right relationships and had the chutzpah to propose big plans. It took me a while to understand that the habits of diligent preparation and doing quality work that I’d learned in school were not the only—or even primary—ingredients I needed to become visible and successful within my organization.”  

Mohr’s realization of the importance of network and relationships is an essential point, and one that is illuminated by McKinsey’s “Women in the Workplace” findings—that 90% of new CEOs for the S&P 500 were promoted or hired from similar line roles. It seems a fair hypothesis that outside of founding one’s own company, which several of our documented CEO’s had done singularly or as a co-founder, working one’s way up the ladder and forming the right relationships is key to evolving as a leader.

In a 2012 report that explored the “talent pipeline and gender-diversity practices” amongst Fortune 200 companies (selected based on outlined evidence of their promoting gender diversity), McKinsey found that while more women are making it to mid-level management, less move on to executive roles and C-level positions. One key point of analysis is the need for companies to find ways to retain talent. There is a strong correlation of drop-off in females in leadership roles, especially as women move through child-bearing years.

Not all companies are created equal; some have found ways to attract, keep, and promote more women than others, and it was based on interviews with senior executives from these companies that McKinsey identified best practices for ensuring closure of the gender gap in leadership, which they outline: senior leaders being consciously aware and committed to achieving gender-diversity; integrating this awareness into their talent management processes; measuring progress against goals; and maintaining an ongoing spotlight on the issue.

McKinsey suggests placing a “focus on removing barriers that discourage all but the most resilient women,” a character trait that bears importance for both men and women successful in assuming leadership roles. Echoing this idea in response to one of our questions, StoryStream CSO Janet Bastiman remarked, “I’ve met some of the dinosaurs of business along the way, been patronized at times and even had some colleagues try to demoralize me. If I was held back at a company, I had the tenacity to find a better one where there was better opportunity. “

Barriers to Entry

Clearly, the large gap in women-to-men C-level roles is not limited to companies leveraging or selling AI and ML technologies. While in general there appear to be a higher level of women in management roles in smaller to mid-sized companies, there remains a significant gap across most industries and sub-industries. What might be the potential reasons for this discrepancy?

In McKinsey’s 2012 survey of more than 14,000 employees at 14 companies, they found that fewer women than men had the desire to aspire to the C-suite (18% versus 36%), though most women do desire to advance in their career (69%). McKinsey had previously identified four barriers to women’s advancement, including: structural obstacles (lack of networks and sponsor relationships), lifestyle choices, institutional mind-sets, and individual mind-sets—all of which are deeply intertwined and not easy to root out.

McKinsey emphasizes that different companies in different industries will understandably have varying contexts and starting places in how they seek to achieve gender equality, but that the “…magic begins when leaders achieve real diversity of thought.”

According to a 2013 study by Pew Research, men and women want mostly the same things when it comes to a job: a job they enjoy, job security (slightly more important to women, 37% vs 33%), ability to take time off for family needs, and good benefits. Opportunities for promotion advancements ranked lower for both men and women, at 25% and 22% respectively.

Generational Differences

Demographics matter more when is gender taken out of the picture and age is taken into account i.e. adults with young children versus older children, millennials versus older adults.

In the same 2013 Pew survey, four out of 10 Americans said they wanted to be “the boss” or manager, with millennials as the most ambitious group in this category and giving a 65% positive response. Still, men tend to be more likely across generations to want these leadership positions (70% vs 61% for millennial men and women, and 58% vs 41% for Gen X men and women).

There seems to be relatively strong evidence, based on Pew’s results and other sources, that age and generational differences may play a role in female leadership. As Daniela Braga, CEO of DefinedCrowd Corp. stated in response to one of our questions, “I couldn’t have started this company if I was 10 years younger. Without the experience of working at a big tech corporation as I did at the time, it would’ve been much harder for me to internalize the drivers and dynamics of doing business with large enterprise as we do today.” In this respect, a changing landscape at present may not yet be fully realized, assuming that—the startup phenomenon aside—there is any correlation between more experience and older adults (female and male) eventually moving into C-level roles.

In a separate 2013 Pew study on women in the workplace, 45% of Americans polled stated that they still believed society favors men over women in the workplace. Notably, this same response differed drastically between men and women, with 36% and 53% responding to this same question respectively. Even more notable was the rocketed response amongst millennial-age respondents: millennial women were the loudest voice at 75% in noting that more change needed to occur in gender equality in the workplace, while 57% of millennial men voiced a similar opinion.

Sociocultural Expectations and Structures

It’s important to keep in mind that out of Pew’s random data sampling, only 13% claimed to be “the boss”, defined as already in a manager-level position or higher. This may speak to the general nature and goals of an individual, including overall ambition and ultimate values and life objectives. It seems safe to say that these vary on a wide spectrum across individuals, with a more limited number of people interested in holding C-level positions; however, the core question is whether there are socio-cultural barriers and inherent structures that discourage or prevent, either consciously or subconsciously, females rather than males from progressing and moving into senior leadership positions across industries.

The 2013 Pew study also showed that the percentage of women (and men) aspiring to top management jobs drops once individuals reach their 30s and 40s—from 61% of women in their 20s and early 30s to 41% in their 30s and 40s, compared to 70% and 58% for men respectively. A perhaps not-so-surprising statistic stands out as one causal strand: 51% of women (versus 16% of men) have said that becoming a parent makes it harder to advance their career, and women in general are much more likely to experience work interruption due to family in the role as the dominant “caretaker.”

In the more recent 2016 report by McKinsey (and referenced earlier), an exploration of staff versus line roles shows the number of roles filled by men and women are fairly even, up until about mid-management levels. When we look at the level of senior manager/director and beyond, these numbers start to shift, with more men in line positions (like operational C-level positions) and more women in staff roles that generally require less external pressure and responsibility.

In light of the evidence, one thing does seem clear—if women are to have the same opportunity and choice to realize their full potential in any given field, certain sociocultural structures (including expectations and resources available for having children and childcare) will need to be addressed and new ways of thinking promoted across gender and industry lines.

What are the implications for companies, in the AI and ML space and beyond, with more or less female leaders?

Senior Vice President of TIBCO Software Mark Palmer recently published a poignant perspective on the topic of gender and leadership in the workplace, reflecting on his 2010 experience of attending the World Economic Forum conference, and in particular a Power of the Purse session that featured speakers like Sheryl Sandberg and others. He recalled a couple of main insights that address both the potential quantitative and qualitative ramifications of industries and companies that hire females for management positions:

  • A 2016 report released by the Peterson Institute for International Economics professional services firm EY found that companies with at least a 30% female leadership presence had a six-point gain on net profit margins compared to companies with no top-ranking women.
  • Palmer emphasized the collaborative, multi-voice nature of the discussions that took place during the female-dominated session, comparing them with the more often “screaming-solo, ego-driven” male-dominated sessions that took place during the conference; this suggests a more thoughtful and thorough discussion and approach to solving issues when a more gender-balanced leadership team is employed.

While the latter point is Palmer’s opinion, it’s one worth noting and of particular relevance to implications for companies solving tough problems—something at which many would argue AI and ML companies are at the forefront. It’s analogous to a point made by Flowcast’s CTO Winnie Cheng, one of the women we interviewed for this piece. Cheng stated,

“When you’re talking about deploying a solution (AI), you have to make sure you think about the impact on all the different stakeholders. Maybe I’m a bit biased but I think women are more empathetic to different stakeholders, so we’re able to pick up on more. For example—AI is automating a lot of jobs—so (we have to think) how does that impact the person sitting across the room or whose job is manual now, and what would be the right way of positioning the message for your technology?”

In an oft-referenced 2005 meta-analysis of gender differences, Dr. Janet Hyde largely “debunks” the claim that there are noticeable psychological differences between males and females, a “myth” that she believes can do more harm than good in stigmatizing gender in the workplace. A more recent 2011 study published in Frontiers in Psychology, however, concluded that in many cases, gender does affect accuracy in reading body language, with females more apt to pick up on hostile reactions and “lack of emotional content” in body actions and males better at identifying positive or upbeat body gestures.

There are plenty of studies on the psychological differences between men and women and no doubt many more in the making. The point here is not to choose or even support a “winning” side in a debate. It is interesting, however, to compare personal experiences—such as those of Palmer and Cheng—with related studies and to reflect on the potential impact of our gender-based biases, which undoubtedly shapes leadership bodies and the decisions made in those positions across industries. What these observations may point to is that, just as men have particular physical and psychological experiences that make them distinct individuals, so to do women—both of which encourage a particular approach or way of thinking about solutions to problems.

That being said, having both men and women present in leadership positions seems important if only for the fact that striking a balance provides a more holistic perspective on addressing or solving a given problem/scenario. It seems likely, although one would hope with as much objectivity demonstrated as possible, that women in leadership positions will help influence the evolution of gender makeup in AI and ML.

In an emailed response to one of our questions, Mayfield Robotics’ COO Sarah Osentoski noted, “While there are still fewer women than men in the robotics industry, I think this is changing. At Mayfield, our team is 50% female, and our hope is that we can inspire other companies to hire teams that are gender diverse.” And as Cheng further commented, “I think it’s important to have men and women in the room so we can have the best brainstorming session; we might be able to see things in ways that the other group can’t quite see.”

Eschewing the cliché, do you think it’s important for more women to strive for C-level exec/founding roles in AI-related domains? Why or why not?

[Draft] Women in AI 6

[Draft] Women in AI 8

[Draft] Women in AI 4

[Draft] Women in AI 3Women in Artificial Intelligence – A Visual Study of Leadership Across Industries

The issue of more men than women in AI and machine learning spaces has other potential ramifications. In relation to women directly involved in creating AI software, there’s the argument—as written about by Bloomberg’s Jack Clark—that too-narrow a group can program biases into AI. There’s also the historically-embedded misconception that machine learning is “too difficult” or not interesting for females because it involves advanced math and engineering skills, a major socio-cultural gaffe, as discussed by Kane and Mertz (2012) and others.

This outdated way of thinking is beginning to wane in the U.S., in part due to a stronger emphasis across all levels of education on STEM and STEAM programs, as well as a growing number of successful female role models in related technical fields; however, according to a report put out recently by the The American Association of University Women (AAUW), the number of female computer scientists and engineers has decreased by 9% since 1990. There are certainly other roles in which women can serve as leaders in AI and ML industries, but this area is historically a ripe area for debate. In looking specifically at females in roles that require strong math and engineering skills, we likely won’t see the results of today’s educational and other efforts for another five or 10 years. 

Is there anything to the argument that women tend to be more interested in social and humanitarian type fields? A Quartz article makes the claim that women tend to be more interested in the “higher good” or humanistic side of AI, and that presenting ML as a solution for more socially-steeped world issues—a pertinent application addressed recently by Harvard Business Review—would be beneficial in encouraging more women to enter the field. Interestingly, Affectiva CEO Rana el Kaliouby described a seemingly related motivation in response to co-founding the company:

“These technologies that are designed to interact with humans need emotional intelligence to be effective. Specifically, they need to be able to sense human emotions and then adapt their operation accordingly. Eventually I felt like I had reached a tipping point in academia. I was building prototypes, but never at scale. I saw an opportunity to build something more sustainable that could fundamentally change how we communicate with each other.”

While this is another open question, for women who do choose careers aligned to social fields, there seems to be a misunderstanding of the overlap and integration that often exists with science and what machine learning is capable of helping achieve—a point that could also explain in part the gap of females in leadership roles in AI and ML companies.

Further Discussion

We concluded our study with as many questions as when we started, including several that grew out of our own data collection and analysis. In an effort to help further discussion, we’ve suggested a few areas of diverse but related interest for future study:

  1. In response to our first two research highlights (major gaps in leadership in technology stack and visual/imaging sectors), we wonder how many women are studying in these areas compared to men, how many women are employed by these types of companies, and what this might mean for diversity of leadership in these areas in the future.
  2. A more extensive study of female in C-level roles in the AI and ML industries, looking at a larger data set across a wider range of company sizes, would help validate this initial survey and help determine if there are further correlations, similarities or differences in findings.
  3. Performing a similar study but tracking the evolution of female presence in leadership positions over time would be more than relevant, putting these numbers into much greater context and perspective and also pointing to progress and other patterns in AI and ML. 
  4. A study comparing number of females versus males hired in companies with greater or less levels of females in C-level positions might shed additional light on biases across gender. Do women have the same biased tendencies that we often label men as having, or are there real differences in how each approaches hiring employees, subconsciously or otherwise?
  5. The one we all want but that poses a real challenge: a more in-depth study, likely a series of case studies, that explores how gender-based biases could affect product/service creation in the AI and ML space, and how these might impact social structures, relationships, etc. (a veritable rabbit hole). What is gained and/or lost (if anything) with mostly male versus female leadership teams in comparison to those in which gender leadership is more balanced?

Recommended Resources

Women in AI Related Interviews with Emerj:


Image credit: Dr. Jennifer Newman

Stay Ahead of the AI Curve

Discover the critical AI trends and applications that separate winners from losers in the future of business.

Sign up for the 'AI Advantage' newsletter: