As AI advancements leap-frog, men seem to be taking the lead, while women continue to be an underrepresented and marginalised group. The paradox is that the future of AI cannot be shaped by a single perspective or a uniform group—real innovation thrives on diversity.
The Grace Hopper Celebration India (GHCI) 2024, organised last week by AnitaB.org India, returned to Bengaluru after a five-year hiatus.
As a part of the opening note, Sindhu Gangadharan, managing director of SAP Labs India, highlighted a critical issue: “Women today make just 25% of AI researchers globally, which is a glaring gap that reflects not only underrepresentation but also the missed opportunities of building more inclusive and equitable systems.”
She went on to speak about Ada Lovelace, a 19th-century visionary who conceptualised the first algorithm for a machine, laying the groundwork for modern computing and AI. According to Gangadharan, women have always been at the forefront of technological innovation, even if their contributions have often gone unnoticed.
In a recent interview, Lisa Su, the CEO of AMD, said that there is a pressing need to encourage more women to engage in these high-potential fields. “There are still not enough women [in this domain]. In fact, I certainly agree with that.”
Su also added that addressing the gender gap and amplifying diverse perspectives are necessary to tackle these tough but rewarding challenges.
“Women in tech is great, and I’ll tell you why. The beauty of engineering technology is that it’s actually pretty black and white, whether you are successful or not. So, whether the product works or not, your program works or not, you don’t have any grey area in there,” she said.
“And… it’s independent of whether you’re a male or female.”
Resonating a similar thought, Bo Young Lee, president of AnitaB.org Advisory, said: “Here in India, women make up the majority of STEM graduates. So, women should be making up the majority of the workforce, but it doesn’t look that way. And it’s not because the women aren’t ambitious.”
According to the All India Survey on Higher Education (AISHE), female enrolment in STEM courses has increased from 38.4% in 2014-15 to 42.6% in 2021-22 (provisional).
Moreover, according to PIB, the World Bank gender data accessed in 2023 shows that the female share of STEM graduates as a percentage of tertiary education in India (2018) is 42.7%, while it’s 34% in the US (2016), 38.1% in the UK (2016), 32.1% in Australia (2017), and 27.6% in Germany (2017).
Source:https://www.interface-eu.org/publications/ai-gender-gap#findings
Lee explained that the influence of AI, algorithms and the datafication of the world on daily life often goes unnoticed, including the ways AI might subtly limit the options and opportunities presented to individuals.
“A recent research report shows that large language models (LLMs) reflect the ideology of the people programming them,” said Lee.
However, “Now is the time to change that narrative. It’s time for women to take centre stage in the AI revolution, not as participants, but as leaders,” Gangadharan said.
Commenting on the larger question of the need for more women in AI, a Reddit user said: “It’s not about skillset; it’s about the way ML/AI models are trained. Diversity in the training dataset is currently necessary to reduce bias.”
“Pretty sure they just train on internet content. If that is biased then the model will be biased. The engineers involved aren’t filtering the data. Hiring more women engineers in AI/ML won’t fix the bias in and of itself,” another user highlighted.
Women in AI
“When women lead in AI, we don’t just add diversity; we bring empathy. We challenge the biases, and we design systems that are more inclusive and human-centric,” explained Gangadharan.
She also discussed how women’s leadership in AI can eliminate biases in hiring, performance evaluations, and decision-making. “But for AI to do this effectively, it must be developed by teams that represent the diversity of the world it seeks to serve,” Gangadharan added.
Gangadharan also touched upon the fact that generative AI is poised to significantly impact India’s economy. Projections suggest it could add a cumulative $1.2 to $1.5 trillion to the GDP over the next seven years.
Talking about India’s strong technology policy, Su added, “We have 27,000 people, most of them are engineers. We have them all over the world, including 8000 engineers here in India, which is 25% of our workforce. And we love it. It’s great to have a global workforce.”
Meanwhile, Gangadharan was confident that given India’s strong position as a technology services provider, generative AI could contribute over $100 billion to the economy. “India boasts the highest AI skills penetration, with three times more AI-skilled talent than any other country. Over the past seven years, there has been a 14-fold increase in individuals who are skilled in AI,” she said.
In an effort to advance diversity, equity, and inclusion (DE&I) in technology, GHCI 2024 India witnessed over 7,000+ women and nonbinary technologists from across India.
“It’s about ensuring that the technology of tomorrow serves everyone fairly and ethically. And to achieve this, we must champion women and underrepresented voices in AI, providing them with the platforms, the opportunities and the recognition they deserve,” Gangadharan concluded.
Ritwika Chowdhury, founder & CEO of Unscript.AI, also spoke about historical pioneers like Ada Lovelace, Grace Hopper, and Karen Spärck Jones – women who made foundational contributions to computing and AI.
Chowdhury believes that women’s participation in AI development is crucial because “women represent half of the end-users of AI technology, making their input in development vital.”
“Diverse teams are better at identifying potential biases and blind spots in AI systems,” she added.
Furthermore, Shivani Rai Gupta, Jio’s chief data scientist, pointed out that “women’s participation is vital, not just for their technical expertise but also for the ethical and social impact.” She explained that diverse voices help shape more inclusive and fair AI models, ensuring they serve a broader range of people and address potential biases in the data.
The post Why AI Needs Women Like Never Before appeared first on Analytics India Magazine.
