|
Listen to this story
|
In a report by MIT, a student explains what AI is. He says, “it’s kind of like a baby or a human brain because it has to learn, and it stores and uses that information to figure things out.”
For a 10 yr old to give such an explanation on a vast phenomenon like AI, means that we have come a long way. AI has always been there and whether we know it or not we use it everyday. The place of artificial intelligence (AI) in the future of education is the subject of intense discussion. Fans of the technology argue that schools must adopt it and use it to deliver a more effective educational experience, while critics fear that its adoption will have a number of negative side consequences.
There is no clear consensus regarding which point of view is correct. AI does not have to be a one-size-fits-all solution. As with most technologies, implementing it safely and successfully necessitates a complete comprehension of the advantages and disadvantages. To give us more insights on this we had our monthly Roundtable session.
The session was moderated by Rathnakumar Udayakumar, Product Lead Cloud and AI at Netradyne along with our experienced panelists, Chiranjiv Roy, Vice President – Industry 4.0, Applied AI at Course5i, Deepika Kaushal, Deputy Vice President at Piramal Capital & Housing Finance, Shan Duggatimatad, Data & AI leader- Sr Director at Ascendion, Parikshit Nag, Head of Data & Analytics at Indus OS and Anand K Sundaram, Head Retail Analytics at IDFC First Bank.
Speed up to keep up
AI has been here since the machine was invented. Nothing has changed. The fundamentals have never been changed and it will never be changed in itself. The challenge is that the academic body, especially in emerging countries like India, never paced up with AI, because the last 25 years has always been about software development. Because everybody was focusing on Java. Now, everything can be done by generative AI or ChatGPT.
That’s when you require logic and require data. That’s where the premise comes into picture. The economic institutions, especially in emerging countries, had never been ready to do it. We see a lot of things where academics and corporate need to be together. But it actually has never been due to the diversity we have today. We are the largest, biggest population in India. We don’t even know how deep and diverse the number of people who are passing out are.
– Chiranjiv Roy, Vice President – Industry 4.0, Applied AI at Course5i
No point blaming!!
It’s fantastic how far we have moved, from the information being restricted to the elite part of the society. This relates to whether we will have information based education or transformational education. We are used to having information based education which is generated by 10% of the society, the rest 90% of that information is replicated, duplicated and consumed. But now we are entering into the transformation way of learning things.
The section of society, which was restricted from getting exposed to the new way of learning new things and new trends, is easily accessible now. I don’t see any boundaries or any socio economic conditions. Some people may not have access to some quality education, because we do get an education, but the quality of education also matters and the networking matters, exposure matters. But there’s a different set of challenges to deal with.
– Shan Duggatimatad, Data & AI leader- Sr Director at Ascendion
It will take time
Any transformation does not come easy. And it’s not done on an immediate basis, it takes a certain time to establish. In recent years there is too much excitement in the institute’s and academies. Kids learn mobile apps at the age of eight but know nothing. It’s a transformation stage. And sometimes in transformation, it takes a while to get the right things in place. For every person it’s like a dual ceiling.
– Anand K Sundaram, Head Retail Analytics at IDFC First Bank
The online boom
Analytics is all about predictive modeling. Everybody is fencing that some models will run and they’ll change the world but Analytics is a mix of business plus math plus statistics. These are the basic combinations a person needs to know. From an academy perspective, I think they need to blend all of it together. In case they want to keep a pace, the first thing is to get your basics right for every kid who you are coaching.
Online courses are a little boring, because many times, you don’t have a live instructor. It’s more of a recorded session and if I have a question, I don’t have anybody to talk to. Similarly, it goes with Academia right now in India, it’s not really a knowledge gaining kind of institution. Analytics is an upcoming area, if you get certified you get a job, and people come to the job and they struggle. We don’t really look at knowledge, we don’t really teach people how to survive in our environment. Especially when you have to deliver outcomes, because we are in the business of doing so.
– Deepika Kaushal, Deputy Vice President at Piramal Capital & Housing Finance
The amalgamation of valuable and viable
Companies come in and tell you that if you learn to code, as a six year old, you will become the next Steve Jobs, but you probably won’t. And even if you do become the next Steve Jobs, it’s probably not going to be because he picked up a course as a six year old. The rate at which AI is evolving, I think for any student who actually has exposure to all of these three subjects(math, stats and business) to begin with, and formed a strong base there is way more critical as compared to learning about NLP or large language model, some course or learning about decision trees and fitting them in the Titanic data set.
Institutions use some of the open source datasets which are available, and that doesn’t make sense because they come back and talk about what they have done. But if you ask them how the decision tree works, they’ll just falter. If you think about putting AI as a course right now, it’s evolving very fast. Standardising it into the curriculum doesn’t make sense at this point in time. One thing, which we tremendously lack in India, is industry exposure. And there is a gap on both sides. If you look at Academia, they don’t have enough industrial projects going for them.
In the real world scenario, if you look at most of the corporations, they don’t have a strong AI or r&d division, which basically only focuses on research. Most of the AI teams, in corporates, have some business goals, which they need to achieve. And they need to deliver on that within a particular timeline. Because AI is expensive. I think that gap needs to be bridged really fast.
– Parikshit Nag, Head of Data & Analytics at Indus OS
Education system needs to adopt
There are two aspects to it, one is the creation of it, and then governing it. Education is not just about technology itself. I don’t think we are there yet. None of the institute’s are really wrapping their heads around. What do we teach our kids? How do you make it easily accessible and easily comprehensible by our school kids or college kids? More than creation, the biggest challenge, the academy still needs to think about, is to educate or create awareness about how we govern this.
Some people refer to it as the curse of magic. So technology is not something that happened yesterday, it is just that awareness is exploding. So the technology itself is not really that complicated. We can train our people, we can put them through and grind them out. But how do we harness the power that needs to be educated along with creation of it? None of the institute’s are ready for it yet.
– Shan Duggatimatad, Data & AI leader- Sr Director at Ascendion
Ethics and Compliance
There is no standardized framework, yet in India, or even globally, per se, but it has started to seep into the ecosystem. These kinds of regulations are quite exhausted and have started seeping into the culture of individuals and institute’s as well. The second thing is around data literacy. Why do we actually need to classify data? The thing is, it’s very easy to store data somewhere in a data lake and make it accessible to everyone.
But should the data that’s available be accessible? Or should everyone be able to access, is a very critical question. The point is around data classification, ensuring that PII data is masked and kept separately, it’s not accessible to everyone, people who actually have permission, and it’s all time bound to ensure that they delete the data after a certain point in time. It’s a good practice, but it will take time to actually standardize this into a particular bill. I think it will take some time for it to standardize, but I think it will come out more around ethical AI.
It’s going to be more of an individual responsibility to say that let’s not breach privacy, let’s ensure that we’re doing things and it will be the responsibility of organizations and institutes who actually inculcate this kind of thinking.
– Parikshit Nag, Head of Data & Analytics at Indus OS
It’s exciting to see how the two industries can combine and what opportunities they bring out.
To conclude, the only thing better than learning from your own mistakes is asking a machine learning algorithm to do it for you. – Rathnakumar Udayakumar, Product Lead Cloud and AI at Netradyne
This article is written by a member of the AIM Leaders Council. AIM Leaders Council is an invitation-only forum of senior executives in the Data Science and Analytics industry. To check if you are eligible for a membership, please fill out the form here
