Female business leaders are playing a vital role in AI’s development, safety and social impact. Yet they remain a stark minority in AI fields, representing just 26% of analytics and AI job positions and authoring 14% of AI research papers.
Ironically, we are about to see AI transform many aspects of life that have traditionally been associated with women. From educating our children (the pandemic she-cession was a harsh reminder of women’s outsized role here), to caring for the vulnerable, and managing the household.
AI will soon drastically change how 50% of our population spends their time, and the AI sector should reflect that reality. Yet gender bias can occur at all stages of AI development, from the coding to the training data to user input.
I’ll explore why female involvement in AI development is vital, and the sub sectors that will emerge with this new technological evolution.
Women building for predominantly female sectors
On a recent trip to London, I was inspired by the female founder of AI family assistant Aurora First, which helps manage home and family responsibilities. With much of the discussion around AI deployment focusing on productivity at work, little attention has been given to the ways it can disrupt day-to-day lives of a huge share of women.
What Aurora does struck me as custom-made for the lifestyle and responsibilities of many women. built with the knowledge that can only come from lived experience. Its AI companion slots itself in to help people manage family activities, communications, appointments and more. I believe we’ll soon start seeing the emergence of similar apps that use AI to manage our doctors’ appointments, schedule meetings with teachers, organize our weekly shop, and help us pre-screen, hire and manage nannies.
Women often assume the role of caregivers and workers or entrepreneurs, and simply don’t have the headspace to keep all our ducks in a row. A 2022 study found that women in the US spend 2x as much time in unpaid caregiving tasks compared to men, amounting to four work weeks a year.
If our kids go on vacation, we need to make sure their bag is packed with meds and other supplies. We need to make sure that we’ve bought them first. We need to organize travel logistics. Make sure they have travel insurance. A new wave of multifunctional apps could take some of this off our hands, potentially taking on half the work we need to do as family life organizers.
But this will only work if we have the right people at the helm – people who understand women’s daily responsibilities and can foresee potential risks that may come with these AI solutions.
If a product is designed exclusively by men, it may not account for predominantly female issues. Women represent only 1 in 4 leadership positions in the 20 largest global tech companies – it’s unsurprising, then, that some of the negative repercussions of emerging tech hit women the hardest. If we take the social media industry as an example, Facebook, Twitter, Reddit, Instagram, and Snapchat were all founded by exclusively male teams – and women are 3x more likely to report online sexual harassment.
Female health may get the attention it deserves
The exclusion of women and minorities from “scientific” research is a tale as old as time. The FDA explicitly excluded women of reproductive age from clinical research trials in 1977 – a policy that was only reversed in 1993.
To this day, even when it comes to ailments that predominantly affect women, research often fails to focus specifically on women and how they react differently to men.
Time has helped reduce this marginalization of women, and now, AI may cause us to take a huge leap forward in our exploration and understanding of female health.
A new study by FemTech Analytics mapped 170 femtech companies leveraging AI in women’s health, pregnancy, longevity and more. It mentions AI tools that help track and predict fertility, detect breast cancer, prevent pregnancy complications, and carry out gynecological imaging.
This emerging sector could not only improve women’s health, it could usher in more testing and scientific research specific to the female population. We need women to even conceptualize such solutions in the first place. That means putting them in a position to do so, with equitable access to financing, research and resources.
Subverted stereotypes
Just because some of the aforementioned fields – like childcare and the home – have historically been female-dominated, it doesn’t mean they need to stay that way. AI could open up the door to a society-wide mindset shift … or, done the wrong way, it could engrain certain stereotypes even deeper.
Take the emergence of personal technologies over the past few decades. At-home virtual assistants like Alexa and Siri have been largely feminized – and subsequently insulted by users – which developers later tried to correct for. Humanoid robots have often been hypersexualized. Just recently, OpenAI’s controversial female chatbot voice Sky was described as flirtatious and intentionally “empathetic and compliant.”
Observers talk about how generative AI doesn’t just reproduce stereotypes, it actually exacerbates and amplifies them. A UNESCO report also warned about how gender stereotypes risk being encoded into and even shaped by AI tech.
Founders must be thinking about the long term impact of their AI product on the world and on the perception of gender roles – not implying that certain roles are only suitable for women, or that women are unsuitable for certain tasks. Women are more likely to be sensitive to this need and, crucially, able to do something about it if they approach the issue from a leadership position rather than one of subordination.
An age old problem
The exclusion of women and other minorities from the tech sector is above all a systemic problem that needs far more attention from academic institutions and legislators.
The tech industry has traditionally self-selected for men. Around the time the internet was taking shape, supposedly “scientific” studies associated male characteristics with the tech persona – a false stereotype that still remains to this day.
Our long-held internal biases not only stop women from being considered for certain jobs or for funding, but they may discourage women from entering the field altogether. Just consider that in 1990 the proportion of females in computer and math professions was 35%, and that had fallen to 26% by 2013.
We can’t allow that to happen with the emerging AI discipline. Each company can take steps to undermine the inequalities that divide us – such as selecting job candidates for neutral or predominantly female characteristics – and ensure broader participation in this world-changing technology.
All stakeholders in AI have a responsibility to not allow today’s inequalities to infiltrate tomorrow’s tech, especially as the next generation of companies begin to redefine our daily lives. We shouldn’t have to sing the praises of women to get equal representation in this critical industry, we’re simply necessary – as leaders, researchers, developers and users – to create products that are truly usable by society.
Credit: Source link
