娇色导航

Hello. We're here at IDC

60th Annual Directions Conference here in San Jose, California. And I'm delighted to be joined by Marlanna Bozicevich, who's one of IDC's analysts. She's a research analyst around data platforms. And we're here to talk all things. AI, Marlanna, thank you for joining us.

And so let's start with a bit of a level set, okay.

You're talking to two people who are working in it all the time. I'm giving you an expertise I'd love to hear in a sentence or two what is the state of play with AI in the it by as you were talking to? What's top of mind for that?

Yeah. Well, in the data management space it's really revolved around three key themes.

The first is productivity. How can we boost data management workflows using more AI automation? And the second is really responsible AI. Thinking about the data that you're feeding into these AI models, making sure that that's cleaned and properly governed.

And that really entails knowing the intelligence about your data. And then the third is, of course, a gentle AI, right?

It's the next phase of AI maturity. And it's really to bring in more autonomous action into your data management workflows. And it's where real time data becomes absolutely crucial. Okay. So it's almost like we've got kind of things we can do today to kind of generally improve things.

Then there's insights that we can generate from data. And then there's the genetic piece which is like all bets are off.

It's kind of the future piece.

So if we roll right back to the beginning there, like AI is used to improve data quality and obtain trusted data, you know, in what ways is AI been used to improve data processes, and what was it been used to boost productivity in those data teams?

And how is this impacting other parts of organizations? Yeah, absolutely.

There's kind of two angles to hit at productivity. The first is AI for data.

And that's really centered around improving the processes for your data teams, making sure you can automate repetitive tasks like data cleansing, data tagging, auto generating descriptions, things that really improve the lives of your data engineers, your data stewards, your data scientists, really generating that clean golden record of data.

And then the second angle is really data for AI. How can we now use that that cleansed golden record of data to further boost productivity.

And this is where we can boost the productivity of our business personas and look at creating an AI assistant or an AI agent that then helps increase the democratization of data and promote data access to your business.

Analysts or even executive leadership, and really creating almost a data culture and having those eyes and line of business folks very much connected and approved. It just sits there. Okay. Yeah. So it's almost like a linear like process of building from, from, from the bottom to the top.

So we talked about managing data within an organization to drive efficiency.

How can it buyers use that now efficiently manage data to power AI workloads? Absolutely. We talk a lot about data products. And this is treating your data as a product, not in the way that you'll sell it to an external marketplace. Right.

This is to be used internally to drive more business value. And data products are kind of composed of three major themes.

They may differ on definition from organization or organization, but it's really centered around accessibility, making sure data products are accessible, reusable, discoverable. And the second is ownership tracking ownership of data from its initial curation to maintenance to nurturing that throughout its entire lifecycle.

And then the third is data products must have a business value attached to them, right?

There needs to be a business objective related back to that, prepackaged data and data products, really, they simplify the use of data, making sure that you're connecting your data producers with your data consumers. And they're crucial for AI initiatives because they really facilitate easy data exchange.

When you have that foundational packaged layer, right, you have your data products that connect to your data and your business value together.

And that can be easily used in AI agents to facilitate data exchange. And it gives that the agent the proper context it needs to achieve autonomous action. Okay.

So there's a lot to be done then that would suggest and as we've touched on previously, data and then management of an access to data is critical for a successful implementation of AI projects.

That's what you just articulated. So what do organizations need to do in order to be fully AI ready? Yeah, well, there's a lot of things organizations can do, obviously. But the first kind of layer that we talk about is creating an enterprise intelligence architecture.

And this is really composed of four major points. So the first is your data plane making sure you have an organized framework for your data locations, data sources, data types.

All of these things kind of managed together to tackle your modern data problems. And then the second is your data control plane. This is a the intelligence about your data, making sure you can inform policy for data quality, data governance, data sharing.

And then the third plane is our data synthesis plane that says all your business activity kind of being looped together.

It's really where you see, data and models separately, not as valuable, but when you put them together and synthesize that, that's where the true action and insight comes. And the last one is this business activity, right.

Looping back to those business objectives and making sure the proper context surrounds it. So these four planes really work together to cultivate the proper data culture that's needed, in order to achieve your actual. Okay, amazing.

So I there's a lot to unpack there and we could probably talk for hours. But maybe let's just wrap with, you know, if you had one piece of advice to give to an IT buyer who's watching, watching this clip, what would that be from a data perspective?

Yeah, I would say, remember that your AI strategy and your data strategy are heavily connected, right?

Whenever organizations are focusing on their AI strategy, one of the top areas to focus on is data management. It's always in the top three, and it's really that foundational layer that sets up success for AI initiatives.

And it's all about maximizing the value of your data, whether it's structured, unstructured, real time. All of these come together to to really boost that productivity. Okay.

Wow. Amazing. Great advice. As I say, we could we could speak forever, but I really appreciate you giving us this time. Marlanna, thank you so much. Absolutely.

Thank you