Wisdom From The Women Leading The AI Industry, With Kavita Ganesan Of Opinosis Analytics

Given how women have been marginalized in the tech industry in general, we try to fit into molds set by our predominantly male bosses and colleagues. This makes us blend in, but what we really need is to stand out for our unique strengths. Women often bring a different perspective to the table. So, I would advise women to be unapologetically themselves (in a good way). Show up every day with their authentic selves, and give their team or clients what they have to bring to the table. It’s easy to say don’t worry about being judged, but if people are shunning you for your unique strengths, that might not be the group of people you want to be around.

As part of our series about the women leading the Artificial Intelligence industry, I had the pleasure of interviewing Kavita Ganesan.

Kavita Ganesan, author of The Business Case for AI: A Leader’s Guide to AI Strategies, Best Practices & Real-World Applications, is an AI advisor, strategist, educator, and founder of Opinosis Analytics. She works with senior management and teams across the enterprise to help them get results from AI. With over 15 years of experience, Kavita has scaled and delivered multiple successful AI initiatives for Fortune 500 companies as well as smaller organizations. She has also helped leaders and practitioners around the world through her blog posts, coaching sessions, and open-source tools.

Thank you so much for joining us in this interview series! Can you share with us the ‘backstory” of how you decided to pursue this career path in AI?

It all started with me being at the right place, at the wrong time, which snowballed into an exciting career. Let me get specific with that. I did my master’s at the University of Southern California (USC) around 2004, and USC had (and still has) a lot of cutting-edge research going on in AI. I took several AI classes under world-renowned professors and also did some research work with some of them. I was blown away by how fascinating the field was, and I wanted more — I wanted to apply AI to real-world problems. However, at that particular time, data science and AI were not really a thing in the commercial world. The only way to “use AI” was to become a research scientist at a big company, and for that, you needed a Ph.D. (and I didn’t have one at the time).

That’s when I decided that I was going to pursue a Ph.D. in the field. After working at eBay for a few years, I ended up joining the Ph.D. program at the University of Illinois at Urbana Champaign and specialized in the intersecting areas of applied NLP, text mining, machine learning, and search technologies. My goal was to become a research scientist when I graduated.

As luck would have it, as I was about to graduate, big data and data science was an up-and-coming thing. So, instead of joining a research lab, I just went on to become an industry data scientist, and at the same time, companies approached me for help in implementing AI, which is how my consulting business got its start. Since then, I’ve been in the weeds on the implementation side, I taught technical people, educated nontechnical people, and now, I work with leaders and mid-managers to successfully implement AI in their organization. I teach, advise, fix organizational challenges, data challenges, and more.

What lessons can others learn from your story?

While it may seem like I predicted the future, I actually didn’t. What I was really pursuing was a career that I’d be extremely satisfied doing and one that would seem more like play than work. This is how I really feel about it now — the intersection of AI, entrepreneurship, and helping businesses get value from AI truly excites me.

None of us are able to achieve success without some help along the way. Is there a particular person who you are grateful towards who helped get you to where you are? Can you share a story about that?

I’d say it starts with my parents. I grew up in an environment where I was never told I couldn’t achieve great things or couldn’t pursue what I thought was best for me. Plus, academically, my parents always wanted their children to realize their full potential — there were clear expectations. My mom also always taught me to “put others first,” which taught me to give, and not just take.

The values I grew up with now naturally drives me to pursue the things that I really want to pursue. I have my own take on what success looks like for myself, and that’s what I strive towards every day. And for me, success is a combination of what I’m able to achieve in my career, my family life, and the ability to help others along the way.

What are the 5 things that most excite you about the AI industry? Why?

The five things that excite me about AI as an industry:

1. No one problem is the same. Even though I work on *similar* problems for clients, the AI issues are not one and the same. The organizational constraints, budget issues, data, available resources, and use cases are all slightly different with each company, making my work challenging yet interesting. I’m a problem solver, so being able to solve different types of problems for a range of clients is intellectually very stimulating. I learn every single day.
2. Power to truly create value for businesses, and help society. Although you hear all these scary things about AI, from a practical standpoint, it’s a really powerful tool to enhance human lives and keep us safe. One story that makes me smile is, during the peak of the pandemic, a hospital in Bengaluru in India used two AI robots to triage patients at a time when vaccines were unavailable. The robots were used as a buffer between doctors and patients at a time when social distancing was critical, allowing doctors to be safe and patients to get the care that they needed. The robots were used to “see”, “understand,” and “communicate” with human patients. This shows you how AI-driven tech can be used in truly effective ways.
3. Ever-evolving techniques. While AI as a field has been around since the 1940s, the techniques are constantly evolving. Some old methods are re-released with a new spin on them. Or techniques that were not scalable years ago can suddenly solve complex problems because of the available computing power. Every few months, you read about this and that technique in the media. It keeps practitioners on their feet as you need to be constantly “running,” which keeps things really exciting.
4. The field is vast. AI as a field is not just one thing. It’s a coming together of different disciplines, and there are many subfields in AI. Plus, if you’re in the business world, that adds an additional dimension. You can focus on the technical side or the non-technical side. You can focus on the tools or the techniques. This again makes the field super-welcoming of people with eclectic backgrounds.
5. Tools are making practical deployments faster and easier. When I started doing AI way back in 2005, AI tools were close to non-existent. The few tools that we had were more like research prototypes only suitable for educational purposes. They were archaic, to say the least. We had to pretty much code everything from scratch if you were looking at any sort of practical deployment. Around 2011, with the rise of big data, the growth of AI tools accelerated. Today, AI is becoming much easier to develop and implement, and I don’t see any signs of this slowing down. In fact, we’re trying to democratize AI at such great levels that low-code and no-code platforms are starting to gain popularity.

What are 4 things that concern you about the AI industry? Why?

1. Misunderstanding of what AI really is at the leadership level. Leaders and practitioners see AI very differently. This is one of my biggest concerns because there are often mismatched expectations between leadership teams and development teams leading to numerous shelved initiatives. When leaders and practitioners don’t see eye to eye on AI, then leaders are less likely to fund initiatives and the adoption of AI can be much slower. This is one of the biggest reasons that prompted me to write my book, The Business Case For AI.
2. Over exaggeration of AI capabilities in the media. AI techniques are constantly evolving, and there are research papers published almost every day on incremental improvements as well as breakthrough ideas. Unfortunately, when the media gets a hold of these research papers, they publish them as though these technologies are ready for prime time. Most often than not, they’re not. AI research work is often tested on a small scale and, then too, on benchmark datasets that may not fully portray reality. This is dangerous as people will have the wrong expectations of AI. Supposedly, “it can cure cancer”,” it’s conscious,” “it will take over their jobs”, etc. Most of these turn out to be just sensationalized attention grabbers.
3. Developers chasing shiny new tools and techniques. Unlike buying the latest and greatest iPhone, using the latest and greatest AI techniques can be disappointing for many businesses. Some of the techniques may be too sophisticated to fit in the organization’s existing infrastructure, or they may not scale well or work for the intended tasks. For practical AI problems, the techniques and tools have to fit the bill. But many enthusiasts and practitioners alike overlook this aspect.
4. No distinction between research and practice. Another thing that truly concerns me is that there is no distinction between techniques that are tried and tested and newcomers to the field. The problem with the newcomers is that they’re not well understood, and there aren’t too many resources to fix problems that arise with these tools. Further, some just don’t scale as well as the more classical and widely used approaches, which is important from a practical standpoint. I’ve seen this resulting in countless failed initiatives, which could’ve been avoided if we had attempted to start with a baseline approach.

As you know, there is an ongoing debate between prominent scientists, (personified as a debate between Elon Musk and Mark Zuckerberg,) about whether advanced AI has the future potential to pose a danger to humanity. What is your position about this?

Both Elon Musk and Mark Zuckerberg have strong AI research teams that are steering the field forward, so I’m pretty sure they’re both informed about the capabilities of AI as it stands today. But certain statements like AI is here to take over the human race can be an exaggeration. Let me explain.

AI can certainly be used to do bad things, but as I discuss in my book, that’s more a function of immoral humans than evil AI. Just recently AI was used to portray the Ukrainian president as saying something he really didn’t, using an AI tool called DeepFakes. This type of act can happen in any circumstance, where you train an AI model to do something, and then you use it not for the betterment of society but rather to do things like spread lies, and misinformation.

Having said that, AI is not one conscious bot that can start making arbitrary decisions like us humans at this time and not for this decade. In the coming years, AI systems will become smarter and will be able to *imitate* human reasoning but I don’t think it will be at the scale of true human reasoning. But let’s keep in mind that if you train an AI system to make decisions on a specific task it can very well do it like a human, and can even surpass human accuracy. However, for us to worry about AI becoming completely conscious and planning a “human extinction event,” we first need to see AI behaving, building upon knowledge and reasoning like humans.

Bottom line, while there’s some truth in the debate between Musk and Zuckerberg, we have to keep in mind that the research books have not shown that AI is at least as smart as a five-year-old in a general sense. However, given how AI systems learn, for specific tasks, you can use AI to do pretty bad and unethical things.

What can be done to prevent such concerns from materializing? And what can be done to assure the public that there is nothing to be concerned about?

Given how versatile AI is where it can be applied in any domain and for hundreds of different use cases — good or bad, we absolutely need regulations around the applications for AI. The US government has been in the talks on regulations, but there has been confusion there on what exactly to regulate. I’d most certainly say we need to start with the applications because that’s immediately how people and corporations are going to misuse AI. But for this to work, the regulations have to be specific and should have exceptions. Otherwise, it’s going to cripple innovation, which is not something we want. Also, corporations and government bodies should invest in general AI education to put an end to the hype, misunderstanding, and overstatement of AI capabilities.

How have you used your success to bring goodness to the world? Can you share a story?

One of the best things that I’ve done in my career is starting and maintaining a blog. Through my blog, I’ve not only forged strong friendships, but I’m also able to teach and help underrepresented AI practitioners around the world. For example, just recently, after following my blog for years, someone reached out to me. He’s a young software engineer from a small town in Bangladesh and a very delightful person. He’s not an AI practitioner yet, but has very strong aspirations to become one but doesn’t know the best way to transition from a software engineer to a data scientist in his country. Just through a casual chat, I was able to chart him a path to follow to get him into a data science role in the next few years — no strings attached. I’ve been able to make many such contributions over the course of the years, and some of my blog readers actually come back after several years to tell me all the places they’ve gone. I believe that with my new book, I’ll be able to contribute in bigger ways around the world, as books are said to be “Uniquely portable magic.”

As you know, there are not that many women in your industry. Can you share 3 things that you would advise to other women in the AI space to thrive?

1. Given how women have been marginalized in the tech industry in general, we try to fit into molds set by our predominantly male bosses and colleagues. This makes us blend in, but what we really need is to stand out for our unique strengths. Women often bring a different perspective to the table. So, I would advise women to be unapologetically themselves (in a good way). Show up every day with their authentic selves, and give their team or clients what they have to bring to the table. It’s easy to say don’t worry about being judged, but if people are shunning you for your unique strengths, that might not be the group of people you want to be around.
2. We also need more women in leadership roles so that they attract more women into the company and into similar roles. Unfortunately, this doesn’t just happen; it has to be part of a company’s hiring and promotions and rewards process, where they’re actively recruiting and promoting women within the company. More gender balance in technical roles will normalize the existence of women in technical roles.
3. My third advice to women in AI is to find their niche. As I mentioned earlier, AI as a field is vast. To be recognized for something, you need to find your niche in the field. It can be anything, from a specific technical skill to something more high-level like a master of tools. Both men and women try to be AI generalists, but it’s hard to master it all, and you end up being unrecognizable for any one specific thing. Represent your niche by showing and not just telling. A niche can work wonders when it comes to attracting opportunities that you’ll be truly interested in.

Can you advise what is needed to engage more women into the AI industry?

Let me focus this answer on corporations. I think, for starters, there needs to be more gender balance when it comes to AI roles within the organization, which will make it less intimidating for women to be in AI. Given the talent shortage in the field, one way to get more women into AI roles is for organizations to reskill and upskill women from different job functions so that they have a way to transition into these lucrative roles. For example, in some Silicon Valley companies, you find many women in product management and software QA roles and fewer of them in AI roles. Some of these women are already familiar with computer science fundamentals, but they need the training and the confidence to move into AI roles and do well in them. Further, women are often intimidated by AI as there are all the techniques and jargon in the field, and some of them may not have a strong computational background. But really, that’s ok. The AI field is vast. They don’t just have to be on the implementation side of things. They can focus on the project or product management aspect of AI which again can be a part of the reskilling and upskilling of female employees.

What is your favorite “Life Lesson Quote”? Can you share a story of how that had relevance to your own life?

I used to go by “the impossible is often the untried.” It taught me to pursue what I wanted and not just follow the crowd. Often, the crowd is pursuing the most common thing, where the alternative seems impossible, difficult, or uncool. Through this quote, I taught myself that it doesn’t really matter if it’s cool or not, tried or not, possible or not; if it’s something I think is important enough, I must try it. One example of it in action is when I gave up my Silicon Valley job to pursue my Ph.D. because I really wanted it, and “the crowd” was discouraging me from doing it as some of them thought that I’d be “too poor” for far too long (which wasn’t far from the truth, but oh well). In the end, it turned out alright; I truly enjoyed my research projects, had enough funding to get by, and most important of all I had a great advisor and mentor.

You are a person of great influence. If you could start a movement that would bring the most amount of good to the most amount of people, what would that be? You never know what your idea can trigger. :-)

Hands down, I would say, do great things with consideration for others. In a rush to make money or make a profit, people forget to be considerate. I’ve been in situations where people try to pull others down to get ahead in their careers. Companies use deceiving and downright dishonest marketing tactics to maximize profits. Companies put profit over the obviously negative impacts on society (I don’t have to name names). The victim in all of this is the other person whose being pulled down or deceived, or negatively impacted. I always ask myself, why? Why do we need to hurt another person to get ahead? Can we not live with just a little less, or be just a bit more honest and let the other person also thrive? We’re all in this together, and if we can operate with consideration for others, the world would be a better place. Corporations would take the negative impacts of introducing certain products and services more seriously, individuals would consider how their actions will impact another, businesses would think twice about exaggerated marketing. We just need a little more compassion and consideration for others at every level of our society.

How can our readers follow you on social media?

You can follow me on LinkedIn and Twitter. But the best way to really keep in touch with me is through my mailing list (if you’re interested in Business AI, that is). You can also learn more about me on my personal website at www.kavita-ganesan.com

This was very inspiring. Thank you so much for joining us!