Connect with us

Alumni Speak

‘Fear of Terminator-style takeover by AI is unfounded’

Shikha Verma, an alumna of Apeejay School, Mahavir Marg and a Data Scientist at PayPal says instead, the tangible challenge before us it to eliminate bias in AI and Machine Learning



Shikha is a data scientist at PayPal, India building fraud detection products for medium and large enterprises. She has a Ph.D. in Data Science from the Indian Institute of Management, Ahmedabad (IIM-A) with a strong background in Data Analytics, Machine Learning, and Applied Statistics. She has presented her research at various academic and practitioner conferences like Grace Hopper Celebrations (India), the largest technical conference for women in Asia, the Women in Machine Learning workshop at NeurIPS & ICML, and ACM Conference on Machine Learning and Human-Computer Interaction (2020). She has also served as a visiting faculty for courses on AI, ML, and business analytics across management institutes in India. In a candid interview, Shikha calls for the need for more women role models in STEMM (Science, Technology, Engineering, Mathematics, and Medicine), why there’s no possibility of a Terminator-style robot apocalypse, the need for eliminating AI bias, and more. Edited excerpts:

How to increase and retain more women in STEMM?

Women representation in STEMM is still abysmal. One of the biggest reasons for this is the lack of female role models in STEMM. Many studies have revealed that there is a direct correlation between female role models and an increased passion for STEMM subjects and greater self-confidence among girls and young women. Secondly, women are not encouraged to take risks, thereby missing out on the rewards that can come from taking big chances. Girls are encouraged to pick up professions, such as teaching, that will allow them to spend more time with the family and kids. Gender discrimination reinforce cultural stereotypes about women and their ability to perform in male-dominated STEMM fields. It hampers confidence and causes self-doubt. I also think women are over-mentored and under-sponsored. Talking about myself, since school, I always wanted to implement what I learned to make a big impact. During my first job at ZS Associates, a US based management consulting and professional services firm focusing on consulting, software, and technology, I realised that Data science and Artificial Intelligence have tremendous potential to positively impact all manner of life, from industry to employment to health care and even education. This would also help put myself in the driver’s seat at work. As a result, I went on to pursue a Ph.D. in Data Science from the IIM-A.

You said women are over-mentored and under-sponsored. Please elaborate.

Mentors are key figures who guide you through the working world and their contribution can’t be belittled. However, many more women would succeed if more people, especially men, stepped up with real help, rather than advice. A mentor will talk with you, but a sponsor will talk about you. That’s why we need more sponsors to pull women up through the ranks of a company or industry and help advance them into leadership positions. This would also help shrink the gender wage gap. I also believe that women should focus more on self-promotion. Men are far more at ease with self-promotion than women, which contributes to a broad disparity in promotions and pay.

Many fear a Terminator-style takeover by AI. How genuine are such apprehensions?

People remain concerned about the rise of Artificial Intelligence with some even fearing it can become powerful and take over the world, but I really don’t see that happening. We tend to be afraid of what we don’t understand. Fear of Terminator-style takeover by AI is unfounded, however the real challenge before us is to fix bias in AI and Machine Learning.  While machines are theoretically neutral, there have been cases in recent years that show even algorithms can be biased. Some prejudices held in the real world can creep into algorithms in several ways. AI bias is a phenomenon that occurs when an algorithm produces results that are systemically prejudiced due to preferences or exclusions in training data. The bias could also be due to how data is obtained or how algorithms are designed.  At a time when AI is dominating our lives, being acutely aware of such risks and working to reduce them is an urgent priority. Also, as data is becoming more and more valuable, we need stricter data protection regulations. Unauthorised processing of personal data can cause great harm to individuals and groups.

What’s your advice to students who want to make a career in Artificial Intelligence and Data Science?

Data Science, Machine Learning and Artificial Intelligence are touted to be the next most sought-after fields. I want students to be more inquisitive about technology and go beyond their textbooks to learn about it. For instance, how does your smart watch count steps? Or, how does a mobile touch-screen work? The emphasis should be to learn more about how technology around you works.

Dheeraj Sharma is Asst. Editor (Newsroom). He covers events, webinars, conducts interviews and brings you exciting news snippets. He has over 10 years' of experience in prominent media organizations. He takes pleasure in the small things in life and believes a healthy work-life balance is key to happiness. You can reach him at [email protected]