娇色导航

Welcome to 娇色导航Leadership Live. I'm Lee Rennick, VP tech Evangelist for IDC. And I'm thrilled to welcome Ritu Jyoti GVP GM of AI and data for IDC. Ritu, thank you so much for joining the show today. Welcome. And could you please introduce yourself, Sure. Thank you.

Lee, it's such a pleasure to be here with you, and I'm super excited to share all the research that we at IDC have been doing.

I, as you said, I lead the AI and Automation Data and Analytics market research and advisory services at IDC, and I am very privileged to work across the technology domains, across geographies and across industries and bring in very, very insightful insights to the end users on this very rapidly changing, evolving market state to.

You're going to be on the stage tomorrow presenting on AI and especially eugenic AI. So I want to dive right into this. So it's great to be at IDC directions, and we'll be presenting some research tomorrow on energetic AI and how it's impacting business.

mostly I, I was I'm speaking with are talking about being in full blown mode with AI right now. They're really looking at ways to create increased productivity across the tech stack and for their customers.

Now I wanted to talk about that first, the impact of AI on the enterprise. So some research from IDC projects that, it will have a $22.3 trillion global economic impact by 2030. So I thought we might start off by just talking about that.

Maybe you could let me know how that will impact business. Yeah. We are at a very, very transformative phase of AI right now.

As you and I know, AI is nothing new. We have been talking about AI for ages, but in the last couple of years we have really seen the inflection point as to how it's making a difference. And it certainly became a conversation at the C-suite.

I started doing AI research or leading this practice in 2019, and I was shouting from the roof that AI is coming, but I will always see that people were like doubting it.

Now people are seeing tangible results. You know, as per our research, we have seen that, you know, Gennie AI is making a significant Ottawa for an average of $1 that is being spent. People are realizing almost 3.5 to 3.76 ROI.

So this study that you're talking about is actually done by a macroeconomic excellence team. And they are predicting that the global economy impact cumulative economic impact of AI will be 22.3 trillion by 2030.

And this will contribute about 3.7% of the global GDP at that time. And it will have four important aspects to it. So one of them is the direct impact, you know, so all the providers that they're selling AI solutions and services.

And then because of that, they will also be providers in the supply chain that they will be providing, you know, networking and server and storage.

So this is the indirect impact. And there's another indirect impact of it is that all the people who are adopting it will have economic stimulus out of it. And the last one that we are talking about is the induced impact.

So induced impact is basically the net new household wealth. Because if incremental new jobs and roles.

Well, I love this part, which it says that for every $1 that is going to be spent on AI solutions and services, which is the direct impact, there will be almost $4.8 of an indirect, you know, ROI on that.

So that's really, really remarkable. So in a way, the research is saying maybe contrary to what sometimes we hear in the media that's going to take away jobs or it won't have as much an economic impact. We're going to see increased economic impact through the use of AI.

So there is an interesting angle and nuance. Okay. And we didn't have this one planned, but yes, yes.

So, I would say that, you know, historical, you know, revolutions in the industry show us that every time technology disrupts, it also creates. So there'll be some massive disruption in certain areas. There'll be certain jobs that will be gone forever. Yeah.

And, you know, in the first era of

automation, the traditional automation, we saw a lot of low level entry level automation happening, whether it was data entry or whether it was, you know, say, a document extraction.

But in this case, what is going to happen is because the automation is going to go to the next level. So I promises to have human beings like you and me. It's the concept of a digital label. It can plan, it can reason and it can iterate.

It can analyze. It's almost like getting a team of people working with you hand in hand.

So, a lot of middle level jobs, which actually has advanced levels of automation, of doing some deep research, doing some, you know, creating some reports, creating some customer service interactions.

They are very poised for massive disruption, including the software development, which is one on the top of the radar. But at the same time, we're also looking into some new jobs being created, like ethicists, you know, account managers.

There's also a fun conversation going on that will have, human manager for the AI agents. Right.

So all right, we are in some fun times, but let's not, kind of, you know, this dismissed the fact that there'll be some massive disruption and the work, the way organizations operate is going to be changed forever. Okay.

Well, this segue is I'm glad we asked that one because it segues really well into the next question, which is about, again, take care.

I will be you'll be presenting tomorrow on our whole conference is really focusing on it. So I'd love to learn a little bit, a bit about agenda AI. You know, first, maybe what are some of the ways that can transform business?

You just mentioned a few, you know, potential opportunities. And then for the technology leader listening in like what are some of the steps they might want to take to look at how they might use agenda AI to transform their business?

Yeah, yeah, the opportunity is endless. I think first, I'd like to dismiss some of the myths in the industry. A lot of organizations are thinking that agent AI is a subset of generative AI. The answer is no. Generate me.

I was a branch of computer science, which was all about creating new content based on existing content. So you can create a new blog, you can create a new video, you can create music, you can create protein structures.

And it's going to be really, really transformative in the way that, you know, it actually enhances the workflow. It enhances humans productivity, human productivity. So suppose

if I were supposed to, put together, you know, a code or an application, I could get the assistance from the GitHub copilot of the world and I could actually improve my productivity.

But when you're talking about a genetic AI, it's basically a subset of machine learning and deep learning techniques that help systems to kind of exhibit agency, set goals, plan. Reason three through perception, reasoning, and loop. It's just like a mini human working like that.

And so and, you know, all this paradigm shift is happening because of all the enhancements, enhancements that we've seen and the, the foundation models and most of them are currently at the text level.

But in the future, we are going to see multimodal ality. You know, it can almost do all the things that we have been doing as a human. And the most important underlying thing is that it's taking autonomous actions on your behalf. It's not just giving you assistance.

You're asking a question by a prompt and it's giving you some insight.

So, now that we have level set that there are two very, very different things and the whole goal is very different.

We are in the very early stages of agent AI and most of the things like I always kind of, communicate when I speak to the end users is that all these years you've been making use of assistants, you know, in the last two years, especially after Genii, and now you want to kind of take it to the next level.

So let's take through an example of the, you had the customer, you know, order status inquiry. So it was an assistant to ask a question. And the response, you get it by, my assistant, which was working 24 seven, can give you your responses and assist you.

But the human was still in the loop and taking the action.

The next stage, which we are at the inflection point right now, is the autonomous AI agent.

So you're kind of thinking about suppose you add as a customer, you actually are looking for, you know, a scenario in which you want to actually request for some kind of a refund or exchange.

A very interesting example that comes to mind is that someone bought an outfit for a wedding, and it does not have the right size, and he or she actually wanted to change it and get it the refund.

Sorry, the exchange within a stipulated amount of time, right? But when they went online and requested for it, the still stuff was not available. But the assistant would have just responded by saying that it's not available and you can actually give you a refund.

If it is an agent, it's reasonable.

It's a good how that, you know, this customer stays in Manhattan and there is a store close to his house, which is actually, having something in the store because somebody just refunded, or returned one of those kind of things, size of the outfit.

So it reasoned and went back to the customer and said, Mr. Customer, what if I get it ready for you in that shop? Will you be able to go and pick it up?

And, it can be ready for you within the next 36 hours for you to pick up. And the customer responded, yes, that would work out.

Then the agent also reasoned, because of all the inconvenience, just to be a customer service

person is doing right, that we can actually offer you a refund, sorry for the inconvenience and give you a refund. And this case, you know, you could still have a human in the loop who can decide the amount that has to be refunded.

But with more advancements and trust in technology, we can actually see a fully autonomous agent.

So what my advice to the end users is that the day is coming when you will have a fully autonomous customer service representative, which cannot just address the level one and level two of the task, but also can do level three, level four and level five.

It can do all kinds of reasoning just the way humans do. So my advice to the end users is that it's early days, but the future belongs to the organizations who can actually adopt it with speed, with precision and trust.

So you get started in this journey, but exercise your due diligence. You don't have to just trip and replace that.

You are using AI and you want to use a gigantic. I don't force that a technology right build the right set of use cases. Focus on the high value use cases. Make sure you're preparing for, people's strategies and the human and machine collaboration.

Make sure you have the right measurement and optimization. I can go on on this, but do you know, kind of policies and experiments then do focused rollout and then kind of prepare to scale but get on this journey.

This is not agent washing going on. This is real. And the better you prepare for yourself and the better you actually focus on the right set of use cases and the decisioning framework, the more you set up for longer term success. That's. Thank you so much.

That's thank you for that overview. And that makes a lot of sense.

And I appreciate you. You go into the detail of how it will probably be rolled out with businesses across the world really right. Okay. So we're just wrapping up the first quarter of the year. It's April 1st today, but happy April Fool's Day.

But I just yeah, you were just talking about, you know, some of the emerging ways, but are there any other emerging areas that you can highlight from your research that tech leaders should be thinking about?

Yeah, I think the, the journey has begun. You know, like in the past decade. And so, today AI is becoming a reality because there's a lot of advancements that happened in the field of data in the field of cloud computing.

And now with all the emergence of the rapid evolutions in the compute powers in terms of edge inferencing, we'll see a lot more explosion of this.

But the next chapter is going to be I know everyone's talking about AGI and people are talking about quantum. Yeah we'll see a lot more. Yes, we will see a lot more, you know, realization of humanoid robotics, physical AI, that's the next chapter that you should be preparing.

That. And what I'm hearing from a lot of the CIOs is they are still going on a data journey.

Some are, you know, very sophisticated in how they're looking at it, whether they're keeping it on prem, on servers or in the cloud. It's a combination now where it's a few years ago, they were saying cloud, cloud, cloud.

Now it's all it's structured and they're doing it specifically to make sure that how they're using the AI is going to be successful. Okay.

Well I appreciate this so much. So as we wrap up the year, I'd love us to let our audience know about any research that, highlights the amount of research you might have planned for the remainder of the year. But, like, yeah, we are going to be super busy.

We are working hand in hand with that and users and technology suppliers as well as investors.

But some of the research that I'd like to highlight is that be on the lookout for the ongoing cadence of surveys, because you'll get a lot of pure insights, and there's nothing better than to learn from your peers. And so what are the best practices that emerging?

The second important part is that we're going to do a lot of talk leadership research on how the stock is going to be disrupted.

So I think, if I'm not mistaken, tonight is the day when we have publishing one of our AI Council research on the disruptive impact of the tech stack, as well as the future of work and services.

And, I'm so, so proud of all the talent that IDC that came together to kind of contribute.

And we are giving some very tangible guidance to the end users for every part of the tech stack, take into account, as well as the technology suppliers as to how they have to play a very, very significant role in helping the end user succeed in that.

Honestly speaking, I think the tech suppliers are also in a very challenging situation because they'll be disrupted the most and they'll have to figure out how do they change their business model going forward.

So there's tons of advice for them as well as to how they can actually kind of capture this momentum and this disruption. In some cases, they'll have to reinvent themselves.

So there's a lot of advice on that. The other important thing that we're super proud of is that, you know, the trust factor that we all talk about is so, so critical.

So we'll be doing AI governance product stage and AI governance market scapes, helping the end users kind of look into what are the right set of tools, which ones should they do?

What are the critical success factors? How should they kind of evaluate these tools. So I can, you know, go on.

But these are some of the most promising research that we are going to kind of, you know, be very, very proud of and, you know, looking at a regular cadence of enhancing them and providing guidance. Fantastic.

I will make sure we include the link in this post so that people can sign up to our newsletters and any information they might need to receive, receive to get involved in the research you're doing, especially as well.

I can't thank you enough. I just want to keep this conversation going and going and going. So I'm hoping that we'll have a test that I'm looking forward to your presentation tomorrow. Thank you so much for joining us.

And I'm hoping we can make this a regular cadence so we can update our listeners. And on everything you're doing in the aspects of research around Gen AI. Absolutely I would love to Lee, and thank you for the opportunity again. Thank you Ritu.

Thank you.