On March 8 every year, International Women’s Day is a global day celebrating women’s social, economic, cultural, and political achievements, and that – of course – includes women in AI.

The tech industry and the AI sector face an ongoing and constant diversity and equality crisis. In a study from the AI Now Institute at New York University, the issue’s breadth is clear.

There is a diversity crisis in the AI sector across gender and race. Recent studies found only 18% of authors at leading AI conferences are women, and more than 80% of AI professors are men. This disparity is extreme in the AI industry: Women comprise only 15% of AI research staff at Facebook and 10% at Google. There is no public data on trans workers or other gender minorities. For black workers, the picture is even worse. For example, only 2.5% of Google’s workforce is black, while Facebook and Microsoft are each at 4%. Given decades of concern and investment to redress this imbalance, the current state of the field is alarming.

While there are systemic reasons for this – starting at an educational level – we can redress the balance by highlighting female role models in AI that the next generation can look up to and be inspired by.

So on March 8, we celebrate and focus on eight leading women in AI whose work in the field is inspiring and moving the entire sector forward.

Mia Shah-Dand, CEO at Lighthouse3 and Founder at Women in AI Ethics

Mia Shah-Dand is the CEO of Lighthouse3, a research, and advisory firm based in Oakland, California. Mia advises large organizations on responsible innovation at scale with new & emerging technologies like Artificial Intelligence (AI).

She brings together diverse groups of stakeholders to build human-centric programs at the intersection of data, technology, and governance. Mia is the founder of the Women in AI Ethics initiative, dedicated to the recognition, recruitment, and empowerment of talented women in this space. She created the first 100 Brilliant Women in AI Ethics list in 2018, which is now published annually, and built the Women in AI Ethics online directory, a resource to help conference organizers and recruiters find diverse talent.

Mia is on the board of the United Nations Association – San Francisco chapter and the Social Impact ABIE Awards Selection Committee at Anita Borg Institute for Women in Technology.

She also hosts monthly events and Twitter chats to raise awareness about ethical issues in AI and highlight the work of BIPOC experts in this space and spoke at Data Natives Unlimited 2020 on the crisis of ethics and diversity in AI.

Timnit Gebru

Timnit Gebru is a computer scientist who works on algorithmic bias and data mining. She is an advocate for diversity in technology and co-founder of Black in AI, a community of black researchers working in artificial intelligence.

Timnit was most recently a Research Scientist in the Ethical AI team at Google and finished her postdoc in the Fairness Accountability Transparency and Ethics (FATE) group at Microsoft Research, New York. On the evening of December 2, 2020, Gebru announced via Twitter that the company had forced her out after highlighting the risks of large language models, which are key to Google’s business.

Before that, she was a Ph.D. student in the Stanford Artificial Intelligence Laboratory, studying computer vision. Her main research interest is in data mining large-scale, publicly available images to gain sociological insight and working on computer vision problems that arise as a result, including scalable annotation of images, fine-grained image recognition, and domain adaptation. As a cofounder of the group Black in AI, she works to increase diversity in the field and reduce the negative impacts of racial bias in training data used for human-centric machine learning models.

Jana Eggers, CEO at Nara Logics

Jana Eggers is CEO of Nara Logics, a neuroscience-based artificial intelligence company focused on turning big data into smart actions. Whether starting and growing companies or leading large organizations within big companies, Jana focuses on understanding the customer needs and creating technology products that customers love and drive business growth. Her software and technology experience comes from technology and executive positions at Intuit (NASDAQ INTU), Blackbaud (NASDAQ BLKB), Lycos, American Airline’s Sabre, and CEO of Spreadshirt. She received her bachelor’s degree in mathematics and computer science at Hendrix College, followed by the graduate school at RPI and supercomputing research at Los Alamos National Laboratory. Her career has taken her from 3-person business beginnings to 50,000-person enterprises.

Valerie Becaert, Director of Research and Scientific Programs at Element AI

Before Element AI, Valerie Becaert was the director of partnerships at the Institute for Data Valorisation (IVADO), which brings together more than 900 scientists to extract economic and societal value from data. She holds a Ph.D. in Chemical Engineering from the Polytechnique Montréal in environmental modelization. Her career began as a researcher in life cycle analysis, a powerful tool employed to evaluate human activities’ potential impact on the environment. Valerie is convinced that our ability to generate, analyze and value big data will change the world and that prosperity and sustainability go hand in hand.

Adelyn Zhou, CMO at Chainlink Labs

Adelyn Zhou is CMO at Chainlink Labs and a bestselling author passionate about the intersection of marketing, automation, and the future of work. She has worked with some of the world’s top companies and fastest-growing startups on growth, blockchain applications, and applied artificial intelligence. She is recognized as a top influencer by Forbes, Entrepreneur, Inc., Wired, Hubspot, and many others. She started her career at the Boston Consulting Group and later led growth efforts at Amazon, Nextdoor, and Eventbrite. She is an internationally recognized speaker at conferences such as SXSW, CES, Inbound, and DLD. She is also an advisor to companies in the SignalFire portfolio, concentrating on growth and marketing initiatives.

Catalina Herrera, Senior Data Scientist at TIBCO

Catalina Herrera completed her MSEE in Advanced Electronic Engineering at Texas University, going on to begin her career as Product Engineer at Texas Instruments. In her career, Catalina has been exposed to state-of-the-art technology solutions across multiple industry verticals, becoming a product evangelist, pre-sales engineer, customer education leader, and data scientist in the SaaS industry. With educational and technical roles to her name, Catalina now works full-time as a Senior Data Scientist at TIBCO, a cloud service provider.

Amy Daali, Founder and CEO at Lucea AI

Having completed her Ph.D. in Electrical Engineering, Amy Daali transitioned between research and engineering positions before moving into a data scientist role at USAA in 2017. Since then, Amy became founder and CEO of Lucea AI, organized the Women in Machine Learning and Data Science collective of San Antonio, and acted as chair for the IEEE Engineering Medicine and Biology Society. Using her experience in Predictive analysis, applied Mathematics, Machine Learning, and Engineering, Amy aims to make healthcare more human, empowering healthcare providers and organizations to make better life-changing decisions through AI.

Terah Lyons, Founding Executive Director at The Partnership on AI

Terah Lyons has been a Technology Policy Fellow at the Mozilla Foundation and was formerly a member of the Board of Directors at the Harvard Alumni Association. Lyons is currently the Founding Executive Director at The Partnership on AI, an organization that aims to establish the best practices on AI technologies and serve as an open platform to discuss AI and its influence on people and society. She previously served as Policy Advisor to the U.S. Chief Technology Officer at the White House Office of Science and Technology Policy. While serving as Policy Advisor, Terah led a policy portfolio in the Obama Administration White House focused on emergent technology related to Machine Intelligence, including Artificial Intelligence, robotics, and more.

Dataconomy celebrates these inspiring women in AI, and thanks them for the work they are all doing to advance the field and share their knowledge and experience with others.

Big Data is no stranger to the women’s movement. In fact, many see it as the great equalizer. It provides cold hard facts, and adds weight to topics that are otherwise easy to skew. Proponents of big data’s use for gender equality range from private firms and individuals to major players, like Hilary Clinton. Even popular magazines have been featuring a recent study on gender inequality among coders. Analyzing polling from 1.5 million GitHub users, researchers at the California Polytechnic University and North Carolina State University found a powerful case of sexism and gender inequality on the platform, and called for the use of big data to highlight the relationship between genders. The paper has yet to even be fully released, yet their numbers and story is blowing up across several news sources.

There are, however, several studies that have been published, and show similar disparaging results. By collecting invoices from corporate legal departments with annual legal spending between one million and one billion dollars, Sky Analytics captured valuable information that thoroughly reflects the industry. The data spans 40,000 attorneys and timekeepers across 3,000 law firms in the U.S. The list even includes 73 of the AmLaw 100 firms, allowing us to peek into top law firms and see what is really happening, right now. The study is also noted as “the first gender study based on $3.4 billion of actual billings,” which is very much thanks to the opportunities afforded by big data. For those in the field, this study may make a world of difference. Research on inequality in the law industry has been going on for decades, but is largely limited to surveys and interviews. Relevant pioneering research is now decades old. This data, however, is both brand new and hard to argue with.

Law isn’t the only field experiencing data-based analysis. Gender Gap Grader was founded specifically to make gender gap estimates on an exact level, using data. They’ve analyzed gender’s role in science, investing and aviation. There may be a common feeling that women are missing in these industries, but these researchers found the exact numbers, making it far harder to argue with or to ignore. For example, to study investors, they used the AngelList database, which includes some 650,000 profiles of entrepreneurs, startup professionals, venture capitalists and business angels, to determine women’s role in the industry. The results? See for yourself.

image source: Gender Gap Grader

Disaggregating Data

The above data is not valuable just because it has been put to use, but because it exists. In fact, missing data is considered to be one of the biggest problems for the gender equality movement. This is what sparked one of the more talked about initiatives of late, Data2X. The joint project of the Clinton Foundation, the United Nations Foundation, the William and Flora Hewlett Foundation, as well as the Bill & Melinda Gates Foundation was founded in 2012 with the clear goal of using big data to help push gender equality. They have been very clear on their “#GenderDataRevolution” mission, and also very creative.

One example of Data2X’s 2015 initiatives includes tracking the mental health of young girls around the world. Recognizing depression as one of the leading illnesses for girls worldwide, they seek to extract information specific to adolescent girls’ mental health. Using Twitter and social media, they hope to gather and disaggregate data and paint a more thorough picture of gender inequality, and how to address it. They hope to use several other methods to generate valuable data, including understanding women’s mobility patterns through mobile phone and satellite data.

The “glass door” of gender does not stop just at women in North America, but continues around the globe, and even broaches several other topics. The International Water Institute worked with the Network of Women Water Professionals and the Women for Water Partnership, to organize the “Gender, Agricultural Water and ‘Big Data’: Practical steps and forwards thinking under the SDGs” meeting Sri Lanka last fall. They discussed the issues of gender and data in the realm of water management and agriculture, and have since released powerful interactive maps of certain areas with sex-disaggregated data. They chart several points, including male and female infant deaths, male and female literacy, and number of workers based on gender. They have also been very explicit in their reasons, bluntly stating:

“Gender inequity is prevalent in water management and agriculture but there is very little accurate data available on gendered aspects of these sectors. In order to improve gender equity in access and decision-making, researchers, policy makers and development investors need better information and understanding of gender relations in water and agricultural systems. The basin maps on this website aim to provide this evidence base for analysis of gender disparities and new development opportunities.”

Is Data The Way Out?

The best way to find the “glass door” of gender inequality is through big data. Though surveys, essays, and even empirical studies are important to the overall progress of social topics, it’s data that makes all the difference. This puts a lot of emphasis on the importance of data scientists and analysts. Companies may gain access to pay rates, but that is not where the information ends. A list of wages cannot reflect just how happy individuals are with their job, or how they may be personally experiencing difficulty. The root causes behind numbers and information may not always be clear. Good data storytellers are necessary to uncover exactly what is going wrong in the pipeline. One cheeky example from Sky Analytics’ legal study includes the slap-in-the-face of which employees control the “large” matters in a company, and which get the “small.” Scenarios involving twenty or more timekeepers, meaning they are more important and involve a lot more staff, were labeled as “large matters.” 93% of large matters were handled by mostly male groups. “Small” matters, on the other hand, were comprised of less than five timekeepers, making seem almost trivial in comparison. 81% of “small” matters were handled by women.

Researchers need data, and they need skilled analysts to determine what that data means. Luckily, with the huge amount of press the topic has been receiving, it seems to be on the right path. It may not happen overnight, but data is slowly filling in the gaps of knowledge on gender inequality, then real, sweeping change may finally be possible.

Like this article? Subscribe to our weekly newsletter to never miss out!

Follow @DataconomyMedia

The common usage cases for big data and employees usually stops at HR and recruiting. However, recent news has sparked interest in the dark uses of data in the workplace. Until recently, the biggest concern about data collection was whether Target would find out you were pregnant, or you would receive obnoxious targeted ads. For all of the positive ways big data can strengthen a company, there are also several downsides.

Employee and Health Tracking For Darker Purposes

It recently came to light that several major companies don’t just track employee health—they know when employees might be getting sick, or are looking to have a baby. Companies like Walmart and J.P. Morgan were cited in a report by the Wall Street Journal for using data to try to save money on employee health. Using third party data companies, businesses can plot data points that indicate what will cost them money, and what will save them money. When it was revealed that 30% of Walmart employees who received second opinions on pricy back surgeries opt not to receive surgery, it became a point of interest. By devising methods to speak with employees about “alternatives” or getting second opinions, Walmart knew they would money. The goal of such data-driven programs is not better well-being, but money saving. Hence why this second example has been splattered all over the press.

Data analysts have also plotted points that indicate when female employees may soon become pregnant. They not only scan insurance claims to locate women who have stopped filling birth-control prescriptions, but they also check for fertility-related search histories. Comparing that data against more basic information like age can determine with great accuracy an impending pregnancy. Questions of “will they fire me if they find out I’m pregnant?” are not uncommon among employees, particularly in the US. Despite existing laws intended to protect pregnant women, knowing that employers are tracking pregnancy might not just be strange, but a massive source of stress. James Hodge, a public health law and ethics professor at the Arizona State University Sandra Day O’Connor College of Law, describes some of the effects this might cause:

“If [an employer] originally thought that 15% of the women in its employee base may become pregnant, but data shows it’s closer to 30%, that could lead an employer to say we cannot hire as many female employees this year because we can’t afford them being out for family leave.”

The true power of big data is not the ability to predict small instances, like one woman becoming pregnant, but trends, like the one described above. Of course, company’s are going to want to use data to get the best information available, and make the best decisions for the business. It is no surprise that initiatives like health tracking to save money now exist. The question is what to do about it. Whether data is being used for the good of the employees or the employers will involve a thin line. Lockheed Martin specializes in a number of areas, including data analytics. While their LM Wisdom brochure sports crime scenes, yellow police tape, and a reference to the Mafia, they have also used their data powers to assist Walmart track employees. The 2012 Black Friday strike at Walmarts in the US apparently did much more than just rattle the cage. Many of the employees were not surprised that they were tracked, but the presence of a company like Lockheed Martin put a very different spin on the situation. In the words of one activist: “We’re artists, not ISIS”

Your Boss May Know You’re Going to Quit Before You Do

Big data is being used to recruit employees that are likely to stay longer. The story of Xerox’s 20% reduction in attrition thanks to algorithm-driven recruiting is the poster child of this method. Unfortunately, the other side of the coin involves analyzing existing employees to find seeds of discontent or signs that they will be leaving soon. One software company called Workday does just that. Their director of data science even told the New York Times, “we’re surprised how accurately we can predict someone will leave a job.” For many, the decision of an employee to quit can seem surprising. It’s a mystery, and there’s only so much one can do to foresee it. Data companies, however, note that there are actually several signs. People tend to leave in “patterns,” and using data to decipher those patterns could save everyone heartache. One positive outcome may be that, the next time you’re planning to walk out, an employer will step in and offer you a raise or better benefits. By comparing what salaries, habits, and needs a person has, maybe data can really help employees. However, when the chief strategy officer of workforce analytics company Visier remarks “we know what benefits they have, we know what they’re claiming in benefits and when,” employees might be left wondering whether that data is going to help them, or hurt them.

One fact employees and employers can agree on is that big data is here to stay. Trying to turn back the clock on how resources are managed and measured would be a nearly impossible task. Even if employees agree to allow parts of their life to be tracked, that does not mean they feel safe with it. Research has shown that employees are wary of large scale, intensive data mining. These kinds of programs lead both to legal and moral questions about privacy, as well as issues in job satisfaction and workplace well-being. Much of the industry is still unregulated in this sense, and employees are highly aware of that fact. Even when an employee chooses not to opt-in to data collection services, data collected from other sources can still be applied to them. Once algorithms are developed, it becomes much easier to turn an employee into a statistic. The onus is laid entirely on companies to show their employees that data is being collected and used in a reasonable way, otherwise the “creepy” factor of workplace data collection might cause major disruption one day.

Like this article? Subscribe to our weekly newsletter to never miss out!

Follow @DataconomyMedia