UPDATED 13:50 EDT / MAY 22 2019

AI

Q&A: Data’s circle of life, from machine learning to AI to world-changers

Yesterday’s static big data lakes have become today’s dynamic data assets. But while data is the fuel that fires the digital economy; it can’t go it alone. Without artificial intelligence and machine learning to bring it to life, data would still be stuck back in the swamp, lurking in the cloud, and gathering dust in the data center.

“Every company is trying to take advantage of new machine learning and AI technologies,” said Anil Chakravarthy (pictured), chief executive officer of Informatica LLC. “[And] one of the key components of making that happen is the availability of the right data.”

Chakravarthy joined John Furrier (@furrier) and Rebecca Knight (@knightrm), co-hosts of theCUBE, SiliconANGLE Media’s mobile livestreaming studio, during the Informatica World event in Las Vegas. They discussed the importance of reliable data sources and strategies for consistent data management (see the full interview with transcript here.) (* Disclosure below.)

[Editor’s note: The following answers have been condensed for clarity.]

Knight: On the main stage, you said that AI and ML need data, but data needs ML and AI. Can you elaborate on that? 

Chakravarthy: Because you have to train these machine-learning algorithms, the data scientists have to be able to find the right data, and then they have to prepare the right data, make sure that they have access to the data, clean it up, and then put it into the AI models, into the AI algorithms and so on.

The training of the algorithms is very sensitive to the quality of the data. It’s really, really “garbage in, garbage out.” If you don’t feed in the right data, the results will be skewed. And so that’s the key part of what we mean when we say, “AI and machine learning needs data.”

The flip side of this is managing the vast complexity and scale of data. Our customers have petabytes of data, thousands of databases, hundreds of thousands of tables. We help them manage all of that.

Data management is not just about availability of data, or the performance of those systems, and so on. All that is super important, but it’s also the security of the data, the governance of the data, the availability of the data to the right users at the right time. Trying to do all that manually, you just can’t keep up, and that’s where you need machine learning and AI to be able to do that for you in an automated manner.

Furrier: You talk about being the Switzerland, the neutral third party, because data needs to connect around multiple sources. You had a lot of industry players up on stage today. How are you continuing to be neutral in the industry as more and more people come in? What does this say about the momentum for Informatica’s strategy?

Chakravarthy: Take any customer, any enterprise, customer, government, any customer of any scale. They’re usually using a lot of different, both on-premise and cloud technology offerings. So, it might be multiple software as a service offerings, multiple public clouds where they’re running platform a service, a lot of different on premise offerings, etc.

All of those offerings that they’re using have a data footprint; and from a customer’s perspective, if they’re using different tools to manage the data for each one of those? Well, they have all the problems they always had — data inconsistency, management problems, etc. If you are a data administrator, are you going to learn four or five different tools to manage the data? That is not really going to work.

So that’s where customers are demanding: ‘Hey, I need a data management platform that can help me manage the data consistently.’ And that’s where we come in. That’s what helps us be the Switzerland of data.

Furrier: This is really about operationalizing AI. … If data is constrained from either infrastructure or regulation, that’s going to slow the feeder concept down. You’ve got to solve that data problem first if you want to scale up operations around AI. What is Informatica doing in this area? And where is the customer’s progress in this new operationalization of AI with data at the heart of it?

Chakravarthy: From an operationalization perspective, first of all you need to help your data scientists and others using AI to find the right data. So that’s the first step; finding the right data. Next is getting access to that data. That’s what you get through data integration, the cloud tools, the big data tools, etc. Then you prepare the data.

We have a number of tools to prepare the data to make sure that the AI and machine-learning models can use them well. Then you feed the data, you run it, you get your results.

Lastly, the explainability is a big deal. Whether it’s regulators or even your own internal executives, they say, ‘Oh, that’s the result of the running the AI model? But how did it come to that decision?’ In financial services, for example, if you’re using AI to make a decision on which customers get a loan or not, you have to make sure that there is no bias in the process. So, in order to explain the results, you need to know where the source data came from. That’s what Informatica does through our governance and lineage.

Furrier: The more enterprises we talk to around digital transformation, the more we hear ‘We want to be consumer-like.’ In other words, they want a SaaS solution, whether it’s an app for banking or an IoT app or whatever, it’s going to need data. How do you architect the data … so that customers have a roadmap to a SaaS solution? 

Chakravarthy: The way we think of it at Informatica is that you have a customer data platform where the last mile of how you reach the customer keeps changing and evolving. That last mile could be through a call center; it could be through a web application; it could be through a mobile app; it could be through a salesperson who’s reaching the customer with a live interaction. It could be a lot of different ways, and it could be all of them. That’s the where the omnichannel comes in.

How you ensure a very good, consistent customer experience is to truly focus on building a customer data platform that can support multiple kinds of last mile when it comes to actually interacting with the customer. Then you take advantage of whatever the latest technology might be. If there are AI-enabled bots, or something else that’s a better way of interacting with the customer, you’re still working off the same consistent customer data platform.

Watch the complete video interview below, and be sure to check out more of SiliconANGLE’s and theCUBE’s coverage of the Informatica World 2019 event. (* Disclosure: TheCUBE is a paid media partner for Informatica World 2019. Neither Informatica LLC, the sponsor for theCUBE’s event coverage, nor other sponsors have editorial control over content on theCUBE or SiliconANGLE.)

Photo: SiliconANGLE

A message from John Furrier, co-founder of SiliconANGLE:

Your vote of support is important to us and it helps us keep the content FREE.

One click below supports our mission to provide free, deep, and relevant content.  

Join our community on YouTube

Join the community that includes more than 15,000 #CubeAlumni experts, including Amazon.com CEO Andy Jassy, Dell Technologies founder and CEO Michael Dell, Intel CEO Pat Gelsinger, and many more luminaries and experts.

“TheCUBE is an important partner to the industry. You guys really are a part of our events and we really appreciate you coming and I know people appreciate the content you create as well” – Andy Jassy

THANK YOU