Agentic AI Services for Hyperscaler AI Technologies with Aishwarya Singh

This is Textron tv. Hey everyone. Welcome back to Textron tv.

My next guest is Ra Sting, uh, ra, or as Ash as she's known, is the SVP of Digital Collaboration Services at NTT Data. Let's welcome her to Text Drunk tv. Hey, Ash, how are you?

Great, Alan, and, uh, great to be here. Thank you. Thank you.

It's a pleasure to have you on here. So, Ash, as I mentioned, you're the SVP of Digital Collaboration Services over at NTT, but you know, you've had quite a career. Uh, one doesn't get to be an SVP at NTT Data, you know, without having some, you know, major accomplishments.

If you wouldn't mind, and I don't mean to embarrass you or anything like that, but maybe share with our audience a little bit about your career. Sure. I started my career, uh, in technology.

That's where I've been, uh, for the last, uh, few decades. I was, uh, designing mobile phone ships at Texas Instruments r and D Center in India. I moved from there into consulting and, um, I was a partner at McKinsey where I was, uh, serving clients, uh, mostly in the technology and tech services space.

And from there I moved to NTD data about, uh, three years ago. And, uh, yeah, it's been a pretty interesting journey. It's taken me to several countries and, and I've really enjoyed it.

Excellent, excellent. Well, Texas Instruments, that's a, that's a name I haven't heard ti I haven't heard in a little while. Um, of course, NTT Data Ash, I think our audience is familiar.

It's one of the largest, uh, uh, technology providers in the world. Uh, digital collaboration being one aspect of it. Um, but we're here today to talk about that NTT data has recently launched some iGen AI services, which of course everybody today is talking Agent ai.

And these are specifically for Hyperscaler AI technologies. Tell us, tell us a little bit about this. Sure.

Um, so NT data, you know, as you know, it's a global technology services and consulting provider. Uh, we do everything from submarine cables and data centers. We are the third biggest data center provider in the world, all the way to actually the, the ID services as well as the, uh, consulting and advisory around it.

Uh, specifically in digital collaboration, we provide all the services that connect employees and clients of large enterprises, which is the employee experience and the client experience services. And the agent AI service that we've just launched is an evolution of it, uh, because we are seeing a huge adoption of Asia Tech ai, both for employee experience as well as, uh, client experience space. We are really excited, we are seeing a lot of traction with clients.

Um, you know, NTT has been on the AI journey for a long time. We actually started r and d in natural language processing about 40 years ago. We launched our AI COE about six years ago, and we've been building and providing gen AI assets over the last few years.

Now with Agent AI services. We've packaged it all in a much more consumable way for our clients. Um, so they can actually use Asian Tech AI as a service.

Uh, and, um, on the back of the key hyperscaler AI technologies. So we've started with Microsoft. Uh, uh, Microsoft is a huge partner for us.

We are actually their leading partner in a number of areas, uh, including teams and telephony. So this has given us an opportunity to truly extend that partnership and use copilot studio, Azure, AI Foundry, and the rest of the Microsoft stack, uh, and provide us services on, on top Ash. They said at the outset, a lot of people are coming out with agent AI kind of services now, and, and agents, we've, this is a topic we've discussed at Text on Gang and at Techstrong and of course our various properties.

To me, the issue is are we all gonna have an army of agents that do our work for us? Are we gonna have one master agent that kinda manages all of these agents? Or is that one agent morph into different, like, is these, is each agent almost an ephemeral throwaway agent that does one thing one time and it's gone, and then the next time we get another spin up, another agent, is it a persistent agent?

How many persistent agents? How do we manage all of these agents? It's a brave new world, right?

A little bit that we're looking at here. How do you and the NJT data look at, you know, humans still have at, at some level, humans, you know, these agents are supposed to serve humankind, right? Like the old movie says how to serve humans.

Um, how, how, how do you think this is gonna work? So, you know, you touch upon a very interesting point because, you know, a lot of, uh, companies still think of, how do I get started? How do I build an agent?

And actually what you're touching upon is, uh, what matters is the whole life cycle. So we think of it as, you know, concept to production and the ongoing management. So you need to think of, okay, here's a technology that can truly change how we work.

Um, I need advice on how to use it. So that's basically, you know, the concept and the design of gen AI and the agent AI services. Then we say, okay, now what do we do?

We need to build these agents. Should I go low-code? Should I go pro code?

Uh, how do I integrate it into the rest of the technology environment? Uh, how do I measure what are the business outcomes that I should expect from this agent that I should already define? And then once the build is done, we get into the ongoing management of it where it's not a static Asian, the Asian needs to continue to perform.

Uh, it needs to remain secure, it needs to be compliant with the internal policies and external regulations, and it needs to, uh, also continue to function even if the rest of the technology environment changes. So we need to think of this whole life cycle, and there are different hierarchies of agents. You touched upon the master agent.

So I think a very basic agent could be a sales agent. So let's say if you're a salesperson and, um, you want to know, okay, what's happening with my client? What's happening with my client's industry?

What's, um, you know, are we having any issues in terms of the delivery of our services that I should know about? What kind of new solutions does NDT data have that I can take to this client today? It can give you all of this information in a very natural, interactive way.

So that's a very basic sales agent, but you could have a master agent that talks to all the different systems, uh, that can even provide coaching services to the, the sales, uh, team. So let's say I'm going to meet a client, I can do a mock pitch, uh, in front of the agent, and the agent could tell me, okay, that went well. Here's how you can do better.

So it's a very interactive way, and it can help at, at multiple levels. So really, I think the agents can and truly change how we do things. I don't have a doubt in my mind they're gonna change how we do things.

I, I just, I guess what I'm having hard time wrapping my head around is how many of these agents are we going to be able to manage? How many, you know, if I have an agent for every, a different agent, for every task I do, for every pot, I stick my hand into it. I, I, you know, at some level, are we better off doing it ourselves?

Right? I mean, there are some things that I think the agents are going to do much better and at much greater scale and efficiency than human ever could. Other things I worry about that we're just gonna feed that capability to the agents and lose the capability of doing it ourselves as well.

Right? I I, we just did a, a, a thing about AI agents and AI in video production, right? 41% of videos today that we're seeing on the internet mm-hmm.

Have either AI in pre-production or post-production, right? It, it's, it's almost replacing your video editors. It's, and that, and then the pre-production, it's writing the scripts, the storyboards, you know, the settings, everything.

And it's great. It's, it's, and it allows, we're, I'm a content right here at Textron, we produce content. So AI and AI agents are allowing us to produce content and publish content and distribute content at a scale we, we couldn't do before.

But I can't help but feel there's, there's a price to pay, right? When you make a deal with the devil, so to speak, there's a price to pay in, in moving that way. Um, do we lose that human touch?

Can, can the, for instance, in content, can these a, can these ais produce compelling as compelling content as humans can, right? We've heard this come from people that, you know, AI has certainly gotten better every week. It gets better, it seems, but yet they say that AI content doesn't perform as well as human written content.

Is is that just temporary? Are our agents gonna get better at that? You know, NTT data is the kind of company that has the wherewithal, the resources to continually improve this to make a difference.

What do you Yeah. Do you think it's possible? So I'll, I'll say a couple of things, right?

Um, I think there is a need for governance in an enterprise, let's say, which is thinking of using Azure and ai, um, to ensure that it's not just a bunch of agents that exist for the purpose of existing, but each one has a clear purpose and people know how to get the maximum from these agents. That's where I think the overall, um, governance comes in. And that's where we also provide the, the managed service, um, around agent ai.

I would equate gen AI to the invention of the internet. It completely revolutionized a lot of industries, how we work and how we do things, and even created new industries, right? Like for example, we are able to have this call because internet came into being.

So I think there is a lot of opportunity that gen AI can create in terms of changing how we work, uh, helping move people away from non-productive tasks and actually using, um, gen AI for certain things that, um, that can help improve employee experience, that can help client experience and also bring in operational, uh, efficiencies. Agreed, agreed enough. Hypotheticals.

Let me turn to real. So the NTT data, uh, Gentech AI services are available now you've launched That's right. How do people engage with that?

So we are, um, though we've launched in late February. Um, the services have been around for quite a few quarters. We already have clients that are undergoing deployment.

Um, we can, um, the way to engage us is, is simple. Um, if you are just getting started, we can help, uh, you design the vision around gen ai. If you have already, um, started and you're on the journey, we can help accelerate what you're doing and we can engage essentially at, at any stage of your agent AI journey.

What we also do is, um, help clients think through what are the areas do they really want to impact, right? So we have clients that talk to us about employee experience, which means, you know, the sales ex agent example that I gave, um, it could be around, um, you know, a marketing function or an HR function. Um, so employee experience related use cases.

It could be around the client experience, uh, related use cases. So, you know, help me if, if I have, let's say a password reset request, can I call in to an agent and get, uh, my password reset done without having to undergo any long, uh, waiting period, right? And the third one is around the operational efficiency.

So if I'm doing a loan processing, can agent KI help me do a credit risk assessment quickly so that I can get the loan processed. So we help plan, uh, think through these use cases and then also deploy and, and manage them. The NTT data website's a big website.

I, I don't expect you to know the URLs off the top of your head, but Ash, how, how do, like where, you know, where within the NTT data website, can we go, can people go to get more information here and, you know, engage more? com/global. And under our services, Egen Care is one of the prominent ones.

I think usually on our website, the front page is all about agent ai. So it should be a direct link. Should be easy to get to then.

Yes, Absolutely. You know, so we, we did our predict, uh, virtual event in January where we talked about, you know, one of the big trends in 2025. And overwhelmingly agent ai right, was the biggest trend for 2025.

And as I said, there's a lot of people making a lot of announcements around agentic ai. I think it's gonna be interesting. I always like to look at those predict predictions a year later and see, you know, how, how true or not true they were.

I, I think this year is gonna be very interesting to see. Now, as you said, N-N-N-T-T data has been doing Agent Agen ai really for a while now, though, in February was kind of the, you know, it all comes together under one thing, but it's gonna be interesting to see how this plays out. 'cause it's game changing kind of of technology.

Um, you know how these things are too. Everybody has overhyped, uh, expectations and then they get disappointed that, you know, it didn't make black into white immediately. So I, I think that's another thing is people need to be realistic and about, about how fast these things roll out, what impact they have, what effects they have.

Gotta give this, you know, you gotta give the wine a chance to, to, to, uh, become wine from grape juice, right? You gotta let it make, and, and it's gonna be interesting to see how it comes out. No, absolutely.

And I think that's why I think before going on this journey, and I really think of it as, as a long journey, I think it's important for enterprises to be clear on what is it that the really going after? Uh, it, is it just because you know that this is the latest fad or is it truly you are seeing it as something that can revolutionize your business? And that's where I think our advisory services really come in, which is, let's start with defining where you want to go, what do you want to achieve?

And then we will take it from there. Absolutely. Ash, thank you for coming on text truck TV today and talking with us a little bit about the Gentech ai.

Right? It's, I think it's the subject we're all kind of grappling with and best of luck to you and NTV data as you continue to roll this out. Thank you.

Thanks a lot, Alan. Great talking to you. All right, great speaking to you.

Ing of SVP Digital Collaboration Services at NTT Data here on Textron tv. We're gonna take a break. We'll be back with more.