Lost Shivaji in a war

Shivaji Dasgupta: don't be afraid of the evil AI

Born in India, did her doctorate in physics in Munich and has specialized in big data and AI in the finance sector for several years. Shiv Dasgupta, Head of Data Architecture & Smart Analytics, has only been with Deutsche Bank for nine months. Nevertheless, he feels at home in the Quartier Zukunft, where we met him. A conversation ensued in which, despite serious issues, one could laugh a lot about the world of thinking data and machines.

To be honest: From a psychological point of view, what fascinates me most about this topic is the fear of people, which is often discussed in connection with AI. Does artificial intelligence sometimes scare you?

No. Why? I see artificial intelligence relatively soberly as a new development. And history has taught us that new technologies always have two sides to it. You can generate energy through nuclear energy, but you can also build bombs. The same is true here too. AI has a lot of potential, such as curing unknown diseases. At the same time, however, there are also negative examples, such as machine-controlled and attackable drones. I would not only want to evaluate these technologies from the fear side, then the positive developments would be lost.

OK. And if we now look rapidly five years ahead, what tasks will artificial intelligence take on in everyday life?

Maybe we can take a step back: What is Artificial Intelligence? From my point of view, artificial intelligence is something that is based on historical data. It recognizes patterns, then automatically provides answers and takes on tasks. When there are very different starting points, such as emotions, it is difficult to replicate them with pattern recognition. But if there are very small standard tasks, such as counting money, it is increasingly automated and taken over by AI. I can well imagine that in the next five to ten years risk assessment tasks in the banking sector will increasingly be taken over by AI.

AI is something that is based on historical data. It recognizes patterns, gives automated answers and takes on tasks.

In many companies, transformation processes are currently taking place with employees who are not yet familiar with digitization. However, I notice a certain inhibition among the employees to deal with it. What do you think is the reason?

The phenomenon exists all over the world. This is because the gap between human understanding and our ability to adapt is widening due to the rapid pace of technical development. People had adapted to the smartphone within seven months. Alexa has been around for a few years, but people can't cope with it because they can't keep up with the adjustment. A person can only absorb change in a linear manner, but technical development is exponential. We have to accept that and focus on further training within the company.

And how could this adaptability be improved?

Optimize training and further education. We are therefore trying to work more closely with leading universities. MIT and Deutsche Bank have that Institute Of Digital Economy founded where concrete thought is given to what modular training formats could be. How can I constantly give my employees impulses so that they look left and right?

It is also up to us that the companies make these new technologies, working methods and communication tools freely accessible to their employees. Slack is a great example. We have now introduced a Deutsche Bank variant of Slack, the Symphony called. You can chat with your colleagues and upload documents.

In an article in the Harvard Business Review you write that some kind of ecosystem should be built out of these data systems. Can you explain that more specifically?

I think the time is over when thinking from the corporate point of view. Today you have to think from the customer's perspective. The customer wants to buy a product and not financing. If the customer wants to buy the product, then I have to ask myself the question: How does he get the product? And how can I get in touch at this point?

If you pursue the idea further, then you are automatically part of the ecosystem. Whether Google, Amazon or Facebook - at the end of the day it's data companies. They live from the fact that they know their customers very well because every customer leaves a data trail. When I search something on Google, I leave my search preferences. When I network with someone on Facebook, I leave my network preferences behind. When I buy something on Amazon, I leave my purchase preference here as well. From this you can then decipher and classify the person very well.

“Whether Google, Amazon or Facebook - at the end of the day it's data companies. They live from the fact that they know their customers very well because everyone leaves a data trail. "

Are we missing out on the boat compared to competing providers?

Between the world's largest platform providers - Google, Amazon, Facebook and Apple - where do we stand? German companies are very small compared to these companies. European companies have to act with one another and politicians should support them. In all corners there is currently a state of shock as we pass all data on to the American or Chinese companies. But who voluntarily gives up access to WhatsApp, Facebook or Google Maps for the European variants? I do not know anyone.

Are there any projects that you are currently working on in Deutsche Bank's IT that will be relevant for end customers?

What we are already doing with data and AI are, for example, these so-called sofa loans. On the basis of other customer data, one defines: The customer is creditworthy.

Then there are other topics, such as the presentation of household bills. Does the customer spend more money on food, clothing or household items? This is already integrated in our Deutsche Bank app and we are constantly improving it with the customer data.

Was there anything, a technical development or innovation in the last five years that surprised you?

The new Facebook currency platform Libra was really surprising. It cooperates very closely with other partners, such as Mastercard and Visa, for payment transactions. Furthermore with Uber and lift for mobility. With Spotify for entertainment - you have a partner for almost every segment. And that's a huge development for existing payment services or banks. We should understand this so that we know what to expect.

The big surprise for Shivaja Dasgupta was the new Facebook currency platform Libra.

And now to finish: is there anything that you are really afraid of?

What am I really afraid of? That we are missing the connection to the new technologies in Europe. I am afraid that the decision-makers in Europe will ask the technology question too late and stick to their old strategies. And that's why we as a company have to do everything we can to be the first to offer this customer service and ask technology questions. I don't think Google or any company is interested in having a monopoly. We gamble away our chances ourselves if we don't act quickly enough and with technical awareness. At the beginning you asked: Am I afraid of AI? No. But I'm afraid that we don't understand enough about AI.

You can find out more about the potential of AI and technology in banking and other areas in our new Smart Contracts compendium.

Claudio Rimmele is editor-in-chief of Qiio magazine. In addition to his psychology and design thinking studies, he co-founded iHeartBerlin.de, which is now one of the most famous lifestyle blogs in Germany. There he regularly publishes articles on art, culture and positive urban development. Together with authors and visual artist, he develops the trend topics and compendia.