Data will put India in a leading position for the age of Artificial Intelligence

I recently had the opportunity to visit Duke University and interact with students at Sanford School of Public Policy and Fuqua School of Business. During this visit I had a wonderful conversation with dean Judith Kelley and recorded a podcast for their Policy 360 series. The focus of our discussion was on Artificial Intelligence (AI) and how companies like Intel play an important role not only in driving these technologies for the data-centric world, but also in collaborating with the government to shape the policies that will enable innovation while safeguarding the interests of people. Reflecting on our talk, here are a few key points that I want to share with all of you.

AI has been around for a long time. Simply put, AI is nothing but the ability of a computer program to carry out a task like a human. Machine Learning (ML), a subset of AI, is the ability to perform a task without explicit instructions, using data to recognize patterns and make predictions or decisions. Traditional statistical models do that as well, but with ML, there is a feedback loop that allows it to learn and improve.

A fast-growing subset within the machine learning category is deep learning (DL). This approach uses layered neural networks that learn from vast amounts of data to solve problems that are difficult to reverse engineer, such as computer vision, speech recognition, and many more. With deep learning, we avoid some of the ‘reverse engineering’ required with traditional machine learning algorithms, instead letting the neural network automatically adjust and adapt to every new piece of training data.

It’s not far fetched to say that AI shapes every aspect of our life in some way and if it’s not monitored, it can be a bit scary. But if we harness AI carefully, the benefits outweigh the risks.

“Corner cases” can make or break AI

When you talk about AI, the US and China are ahead of everyone else in terms of international patents. The good news for India, however, is that AI thrives on data, and we have an abundance of data. Not only the quantity of data, it thrives on the quality, diversity and depth of data. And it fails in corner cases.

A corner case is a rare event. For example, when I lived in Oregon, I was driving to work early one morning and there was a mountain cat that jumped in front of my car and then crossed the road. I hit the brakes and just managed to avoid it. I had no idea what this thing was and it could have easily caused an accident. Now, this is a rare event in the US. But if you look at India, there are cattle, dogs and even people crossing the road at random. So, if we build an autonomous driving algorithm, based on data from this unstructured driving environment, it will be robust enough to deal with these corner cases that are so rare in the US or Germany. I believe that India has an edge when it comes to developing robust AI systems.

Policymakers have their work cut out for them

From an ethical perspective, I think of the AI creation process in three steps – the first is data, the second is the model, and the third is the use case.

With data, we have to value data privacy and build policies to safeguard the interests of the people. We want to then make sure that there are no positive or negative biases in the model. And lastly, the use case. Where are we applying this? We have to ensure that this AI model is being used in an ethical way that makes a positive impact on people’s lives.

Policy intervention is absolutely essential across each stage – in the data, in the model, as well as in the use case. The world of AI touches not only the creators of AI, but everyone else as they will be the consumers. I believe that AI will be so pervasive that everybody will use it as a tool. So, the expectation or responsibility of companies like Intel, is to train people who will be the decision makers, and policy makers on what AI is and what needs to be done before we implement AI.

Data is the differentiator

The wealth of a nation will no longer be determined by natural resources like oil. I believe the most valuable resource will be data – how we create data through digitalization, move or transmit data through new technologies like 5G, how we store it through technologies like 3D XPoint and how we leverage it for AI.

India knows this and for the first time we are seeing the benefit of having 1.35 billion people, because 1.35 billion people are generating data. Our government understands this and is building national level platforms like Aadhaar and Unified Payments Interface which have spurred so many businesses.

Even street vendors today use this payment interface through digital wallets. India was a cash driven economy, and it leapfrogged from cash to digital payment, skipping the credit card in the middle. Similarly, in many such areas, I believe India will leapfrog. For example, people may not have internal combustion engine cars, but they will have electric vehicles. Or they will not have any cars and they will use mobility as a service.

Collaboration is the key to creating AI for humanity

Today, the Indian government is aware that we must put policies in place as we enter the AI age. Industry leaders are part of the process of creating these regulatory frameworks and Intel India has contributed to India’s AI strategy. I am also privileged to lead a 5G consortium that comprises government, as well as industries. On every generation of technology adoption – 2G, 3G, 4G – India has been two to three years late. If we don’t want to be left behind, we need to ensure we understand the needs of India and influence global standards to effectively address them.

As we enter the age of AI, India has an opportunity to lead and leverage our unique strengths to our benefit. We should learn from our early failures and leapfrog to a leadership position. I’d like to end by quoting one of my role models, Abraham Lincoln who had a wonderful philosophy to always “fail upwards”. Growing up, I have had many failures in my life and each one has motivated me to roll up my sleeves and rise even higher. So, let’s all learn from our failures and take a longer stride next time.