Introduction

Good evening, everyone. And I know we started a bit late, but the intention over here is that before we get on to the networking, I just wanted to ask a few questions to you and bring some aspects which would align all of us to use the cool products, something what was demonstrated a while back.

I'm sure a lot of you are already using AI in your day-to-day work. And without going to the Mentimeter, if anyone could indicate how you feel about AI.

How many of you feel excited about it? Nice, nice. How many of you are curious? Amazing.

Yeah, we should always be curious because yes, curiosity is something which will lead us to do better things in life. And anybody who is a little nervous, a little doubtful, yes, that's also true because there are some aspects of AI which should make us nervous. We should make us doubtful. Then only we will be able to bring those guardrails which will make sure that the AI is used in a safe, ethical, and responsible manner.

Leveraging AI Technology

few minutes let's dive into how best can we leverage this technology to transform ourselves our personal lives our professional lives as well as whatever we intend to achieve in terms of our productivity creativity so let's see how best we can use this technology and when we talk to leaders we talk to practitioners like you all they have different things on their mind about AI, how they can leverage it. One of the top most concern is how do they themselves upskill?

So I was speaking to one of our fellow colleagues here who's here right now. They have been a software developer for the last many years and now they are pivoting their career to become an AI expert.

But there are a lot of other people who are domain experts in their respective field. There are doctors, there are engineers, there are architects who are leveraging AI for doing their day-to-day activities. So it's very interesting, but there are concerns which we need to address.

Prominent Figures in AI

So how many of you identify this person? She's very, very famous. She's the godmother of AI.

Anybody would like to name her? Yeah, she is Dr. Fifi Lee. She is a professor at Stanford.

A lot of AI which we use today in our day-to-day life is all thanks to her. So she is the real godmother of AI.

And she is the one who runs a lot of programs which would guide not only the president of the US and other lawmakers across the globe, but also K-12 students how they can leverage AI in a responsible and ethical manner.

The Importance of Upskilling

But what does she has to say about AI? And I would like to bring your attention to this.

It's very important that why all of us should focus on upscaling learning a bit of how ai can help us in doing what we do perhaps it can increase our productivity it can increase our creativity so let's see how we can actually take it forward and this might be appearing a little distressed or little stressful but this is something which we have navigated previously also i'm sure on in this particular hall there are people who have gone through the dot com age who have

seen the mobile smartphone revolution and we have kind of navigated through that a lot of us have changed our profession a lot of us have changed our path of career so i think it's something which is doable as we have done it all our lives how ai is different from what happened in other technologies so i'm sure you must have heard of something called general purpose technology

So what you see on the screen are certain general purpose technologies which have transformed how human beings have evolved over a period of time. It was steam, then electricity, then computers, semiconductors, internet, and now it is artificial intelligence.

I would not be wrong if I say that there will be some sprinkling of AI in everything which we do, right from our mobile phones, to our laptops, to our work, to our own day to day task which we handle something like how we put some salt to everything which we eat except things which are sweet except your coffee or tea but there will be some sprinkling so how much salt has to be added is something which you kind of develop a taste of

Too much of it can spoil the taste. Similarly, too much of AI may be harmful, but then you may require some sprinkling and that calibration is something which is important.

And that's why this technology is something which can be revolutionary. It has a lot of impact in terms of how we provide personalized medication, how we address climate change, how we solve our day to day redundant tasks, which are kind of very boring.

At the same time, it has concerns which can be really detrimental to how we evolve.

Transformative Power of AI

It took human beings almost 75 years to adopt electricity. So it's going to take some time, may not be 75 years, maybe few years, a decade or two, but yes, it's all upon us how do we leverage it.

So there are researchers who talk about something called transformative power of AI. So they compare AI in the current form with something what happened many, many centuries ago in terms of industrial revolution, where the GDP of the world moved up from almost 0.2 to 2%. in a particular rate of growth.

And there are researchers as well as there are studies which are indicating that we could utilize AI, especially generative AI and the future forms of AI, to increase the annual GDP growth to almost 20%.

Last year, I was part of a research which was done over here. And they were trying to find how we can increase productivity of our country, how all of us can be a little more productive. And they concluded, the researchers concluded, that if we are able to leverage and adopt gen AI, we may be able to increase the GDP of the country by 2.3%. in the next couple of years.

There are certain factors where AI has already attained human-level performance. Again, it is human-level performance. It is not human-like. Please do not take me in that sense that I'm saying that AI is human-like. AI cannot be human-like at this stage.

But few levels, like for example, certain translations, certain functions as solving some math problems, AI has attained that level. That means some of your tasks you can offload to AI. and it can perform as much as you would have done it yourself.

But there are issues, there are problems. The current capability of AI is limited to certain pattern recognition, certain creation of text, code, images, video, which will require a human being to verify them and add some more incentive, some more value to it, right?

In addition to this, there are huge limitations in terms of how AI can be safe and responsible. There are issues of bias. There are issues of fairness.

There are issues of AI not being safe and aligned to what you have created the AI for, for which you require human supervision. You require a human in the loop. And that is why it is suggested and the research has also proven that you have to have a combination of human and AI working together.

So my focus of my research currently is to create augmented collective intelligence by how best we can leverage human intelligence and machines to make us do things which we always wanted to do. Let's say so many of us over here wanted to be painters. artists, actors, but you had to do something else as well because you had to pay the bills, right?

So perhaps you could use AI to do a lot of your tasks which are repetitive and you could find some time to do things which actually give you pleasure. There are people who are motivated by doing things for nature. They want to solve climate change problem. They want to serve the larger populace to eradicate poverty or educate people.

There are tools available now. And you would see it, for example, a lecture which is delivered in English in UFT by a professor can be streamed live and a school child who's there somewhere, let's say in Burkina Faso, who does not understand English, he or she can actually watch it, stream it live, understand in their language, and still assimilate what the professor is saying over here. That is the ability of technology which can transform human and human race.

Economic Impact of AI

Now coming to what is the impact on the economy, as I was mentioning earlier, this is a report from McKinsey last year. It says that Gen-AI only has the potential to create value worth $2.6 to $4.3 trillion, which is equivalent to the GDP of UK, sixth largest economy in the world. So it has a lot of impact.

This is one of the lists which was issued last year by McKinsey, 63 use cases against different functions. But it proves, and now we are seeing the result of it, that a lot of companies are leveraging AI to solve certain use cases which are helping them increase the productivity. By increasing the productivity, they are saving costs. And by saving costs, they are enhancing their profits and revenues.

Yes. Jury is still out to see how much is the ROI of GenAI. That will be something which will get measured in the next couple of years as companies, individuals, teams start investing and start reaping the output of what they have invested in a couple of months earlier in terms of upscaling, in terms of developing new models and creating these tools.

Challenges and Concerns

As I was talking about, there are issues in terms of ethical and responsible practices which impact all of us. There are concerns about job displacement. There are concerns about safety and privacy of AI.

There is for sure algorithmic bias, which AI has, because it is intrinsic to the data set which we use. And then there are issues regarding transparency and accountability, which are being addressed in various forums, right from regulation, from guardrails, to practices which are being put in place.

But yes, it is catching up. It is not yet where it should be. It will take a couple of more years perhaps for us to put those practices in place.

As it was brought out earlier, it has been observed that sooner or later, certain aspects of AI will be able to reach human potential of performance. What does that mean to us? That means that a lot of our day-to-day tasks can be done very well by these tools.

It could be like automating a routine task, sending 100 emails. You don't have to actually type and write all the emails. You can have an agent which can perform this task for you.

You can be in the loop. You can check the text and then send it out. It was still happening. People were actually sending bulk emails. But this will be a little more personalized.

And you will see the rise of... agentic AI and multimodal, what Vineet was mentioning. In the next six months, you will have these agents available on your phones and on your laptops through all the service providers, which will actually be able to do things the way you envisage them to be done. And it kind of takes five tasks off your plate for the day.

Bridging the Skill Gap

Now comes very important, the biggest challenge right now, which has been observed by organizations, is the skill gap. Why they're not able to adopt, and this is where the opportunity lies.

There's a skill gap, which all of us sitting here, that's why we are here, right? That's why we didn't go back home and watch a movie or watch Netflix.

We want to focus on that we want to do this gap analysis and bridge this gap. And this skill gap is something which can be filled if you focus on kind of updating and upskilling yourself.

And that is what my call to action would be, that we require the skill gap to be filled at various levels, right from the top senior levels of leadership to the people who are at the operations level and the people who are actually at the functional level. They all need to upskill themselves. And this is something which is moving at a very fast pace.

You did a program today. By the time you graduate and you want to implement something, you want to make a product on it, six months from now there will be a new version of a model which will come from a competitor. So you were building something on LAMA 3.2. Now Mistral has released something which is much cheaper, much faster, more efficient. You will have to shift.

So you'll have to do some kind of changes in your But this is very important for us to align ourselves. And there will be new roles which will keep coming up.

So there are a few roles which perhaps didn't exist in the past.

But they are something which are coming up very, very fast. You will find that today's day, there are more people being hired as AI ethicists, AI governance leaders than AI engineers and data scientists.

So those who are, let's say, who are interested in policy making, they have roles which can be leveraged. You don't have to be a computer engineer or an data scientists to be in AI. AI has a lot of roles which can fit your interest.

So essentially like to ride a Lamborghini right now on blue, you do not need to be an automobile engineer or a Lamborghini engineer. You just need two things, right? Your driving license and a credit card to kind of rent it out. So that is what it is actually.

This is your way to get into the whole AI bandwagon, but in a way where you can actually upscale yourself, learn, explore, and continuously improve on it by deploying and building things.

And like I was talking to someone a while back, each of these professionals are also upscaling themselves. They are the core fundamental roles in AI, data scientists. They have to upscale themselves because things are evolving.

There are data engineers who are kind of doing new certifications, new courses. They are getting their hands dirty with new techniques.

And similarly, the last part is very interesting. Those who are from non-technical background, they could actually perhaps leverage these new developments which are happening across the globe.

Again, don't have to scan it, but how do you think you intend to kind of upskill yourself in next six months? Anybody who is doing some programs, some course, or who wants to kind of learn something new on their own? There are a lot of things happening online as well. Anyone who is doing it or who wants to do it?

I think that's the right call, and you will be very happy with the outcome of it. So just a very, very high level thought. I would suggest that it should be at an organizational level, team level, as well as individually.

We map what we have as our skills, and then we upskill ourselves and continuously improve ourselves so that we move ahead. And it could be that we adapt to the situation, which is evolving much faster. Upskill and use it for innovation.

It could be just that you, instead of asking somebody to make your website, you use something which you learned and created your own website or you created an app for you or your family members or your friends. That is something which is possible to start with.

Ethical and Responsible AI

One thing which I would say would be the bedrock of this is this part, which is generally being left out of consideration, but I would urge all of you to focus on trustworthy, responsible, and ethical AI. And there are enough literature on this, and there is still a lot of gap. Those who are interested perhaps can actually look at this as your new career option as well.

Now there are various projects which are happening. So this is something which a team is doing where they're trying to build different agents which can do anything for anyone.

So let's say you are a solopreneur or you are a single leader who wants to create a project but you don't have a project team yet. So you can use these agents which can actually do a lot of work for you, initial work which can come to you. And then you can kind of review it and then see how it can help you solve the problem which you're trying to solve.

Similarly, one of the very good output of AI is AI for good. So be it education, be it fight against climate change, be it in health care.

So I'm sure you must have heard about AlphaFold. So this time, Nobel Prize, two of the AI researchers, they got Nobel Prize, one for physics, Dr. Jeffrey Hinton and another colleague of his. The person who has been the CEO of Google DeepMind, and they have created a particular application, which is called Google AlphaFold, which has solved a problem called protein folding problem, which has been a problem for the last 50, 60 years.

People have been researching on it, and it was estimated if it was left to humans to continue to solve that problem, it would have taken almost a billion plus years of PhD time. And now, with this particular solution, you could predict the protein folding, and you could have personalized medication for those unfortunate humans who have some genetic challenges.

So that is the transformative power of AI, that we could have personalized medication being delivered to cure our ailments much faster and more effectively.

Conclusion

I have one call to action. So I just want to draw your attention to this.

So as human beings, we can do a lot of things, right? All of us have done it in the past.

We have friends, we have family members who have been doing things more than what they are traditionally being kind of branded as doing, right? So there are chefs who are good musicians also. There are musicians who are good mountaineers also. There are bicyclists who are also good software developers.

So with this, I would urge and quote another very famous AI teacher. He's Professor Andrew Nutt, and he's the one who's the man behind Google Brain, Baidu Brain, and he runs a lot of incentives for educating people about AI, a professor at Stanford as well.

He says, I'm sure there are a lot of people over here who are multilingual. So please add one more language to your plethora of knowledge, and it could be Python. And right now, it's very easy to run Python, because you have a lot of AI tools which can actually help you code. And that brings in a lot of experience and confidence in you.

I know there are a lot of software developers here on the call in the room. But still, those who are uninitiated, you may use this. And the idea is to actually connect the dots between a solution and a problem through all of you. And everybody is important, actually.

AI requires multidisciplinary teams, not only data scientists and AI engineers, but also subject matter experts, AI ethicists, human rights experts, legal teams. So everybody is important and is required. And you will find that these roles coming up very frequently in different job postings very soon.

And if I can be of any help and you would like to continue the conversation, this is my LinkedIn. Feel free to connect. I'd be happy to continue with this conversation. Thank you very much.