The $108 billion Japanese conglomerate NTT (Nippon Telegraph and Telephone) has been expanding global operations. In 2019, all infrastructure-oriented companies came under NTT Ltd. Then, in 2022, it merged all application and digital and consulting companies that NTT Data had with NTT Limited to create NTT Data with $30 billion in revenues, making it one of world's 5 largest global IT services companies. Fortune India spoke with Abhijit Dubey, who will take over as the CEO of NTT Data Inc. in mid-2024 on what this means to people across the world.

How does what NTT offer differ from other similar entities?

Nobody has a portfolio that we do. If you think about enterprise customers transforming to cloud, they're going to build a multi-cloud world, multiple public clouds and a private cloud. Since many public cloud providers sit in our data centers, we can provide for enterprises to be co-located in our data center or the private cloud with their public cloud provider. That is a huge value proposition that no other IT service provider can do.

In manufacturing, mining, ports, airports, warehouses which are seeing a massive convergence of IT and OT (operational technology) environments, once you solve the connectivity problem, especially in terms of ultra-high bandwidth and low latency, then you can unlock a whole bunch of digital use cases. We have built smart data platforms which effectively does real time analytics at the Edge and NTT DATA brings consulting and application knowledge of the applications that reside in the manufacturing shop floor or in the mining environment. It's a unique combination that creates a much stronger value proposition.

Now that Generative AI is around, how do you plan to leverage it and what new services can you offer to clients?

We think about this in terms of three different lenses. In our data center business where we are already effective, and all public cloud providers are in our data centers. This is going to add an incremental capacity requirement. We powered public cloud providers and also want to power the generative AI infrastructure as well. Second, we are implementing all generative AI technologies into our internal operations, processes and have already deployed in some cases at scale which is unusual. Many new services are Gen AI enabled around cloud transformation. We built a Gen AI tool called Dedalow that can take any legacy code and transform it to any new code in an automated way. It's pretty interesting. We have done similar things around Gen AI enabled software development, smart document management, intelligent agents, and also looking at building industry specific large language models as well. That would generate new value for clients and for Generative AI the way we think about it is we will use the generic models that are available in the industry like Chat GPT and Lama etc. But as NTT group we invested $3.6 billion in research and development. We have just recently announced our own lightweight LLM that has a significantly less computational requirement called Tsuzumi. And we will leverage that as well as the currently available in Japanese that will be available in English next year. We will use that as well.

Every country has come with data rules. How do you manage it across nations?

We are pretty tied into most of the important jurisdictions in terms of what's happening from a regulatory framework standpoint. As an example, you know, as Europe goes through Gaia-X, we've been quite involved as well. We know what regulations are coming and what that means for us and we make efforts to do that. We think it's all net positive for us, in terms of business opportunities, The thing that still remains to be seen is what the regulatory framework will come around generative AI. It's still too early and I think it'll take some time.

There is an entire debate in India over private telecom networks. What is your stand on this? How can companies leverage deployment of private networks?

We need private spectrum availability. And, we would be in the camp of having that available in India. There's a whole range of industries. Think of them as the non-carpeted industries, you know if you have an office which is carpeted, you generally have pretty good Wi-Fi and that's good. Good enough connectivity, but if you think about manufacturing shop floors and you know ports and warehouses and those kinds of things, generally non-carpeted don't have very good connectivity. Wi-Fi doesn't work in many of those cases, you do need private networks that are especially very secure, and the solutions we have are generally behind the firewall. It's not just private 5G, but it's also behind the firewall. It is very secure and it's designed for the enterprise effectively as opposed to what the telcos would do, which would just take a slice of the public spectrum, which is not, it's outside the firewalls. It's not very secure.

NTT has been expanding its India presence. Where does that stand?

India is our largest data center footprint in Asia. We have already invested about $800 million during 2021 and 2022, and are going to invest about $500 million annually for the next three to five years. We have 16 data center campuses today. We're going to add another 10 data centers that are under construction. Three of them be live by the end of this fiscal. That's going to add about 88 megawatts of capacity. The data center market is growing globally at around 20% and now with generative AI and high computational infrastructure that AI needs, we think that's going to add another 5 to 7 points of growth in that market. The $500 million goes into data centers themselves, but also into renewable energy firms because we have publicly committed to being 100% renewable energy powered in our data center businesses by 2030. That is a tall order given that we are adding much capacity every year.

It will also go into cable landing stations. There's a new submarine cable that we have built called MIST with 200 terabytes of capacity, which is one of the largest in the world. That connects India, Thailand, Malaysia, Singapore. We have two landing stations in India—in Chennai and Mumbai. And when you land subsea cables in those stations, we can look to build the data centers right about there, which really is very beneficial from a data center standpoint. We are innovating in terms of power consumption to reduce power consumption in these data centers. We piloted liquid immersion cooling, direct to chip liquid cooling in India to begin with.

AI tool Dedalow

NTT will invest $500 million annually for the next 3-5 years in India