NetApp CEO: We Are ‘Solving The Problems For The Era Of Data Intelligence By Bringing AI To Your Data’

‘Without data, the systems are useless. We see this much like what cloud was in the early eras, where cloud was a distinct silo, the on-premises world was a distinct silo, and you needed to bridge the two,’ says NetApp CEO George Kurian.

At The Intersection Of Data And Intelligence

While businesses have been digitizing data and piecing together different data points for analysis, only recently have the tools to get value from that data become available, particularly with the advent of AI, said NetApp CEO George Kurian.

Kurian, speaking Monday at a press conference prior to the kickoff of this week’s NetApp Insight conference in Las Vegas, said the industry is now in what he called the third phase of the modern data revolution.

“Now is the need to go wide in scope of the data, in addition to being long in trend, so that you can fully understand all of the data in your organization so that you can get the best results,” he said. “So, for example, you want to unify the transactional data about your customer together with your most recent videoconference and your customer call center record all into one unified view of your customer. This need to unify all of your data continues to grow in importance.”

Kurian said that the use of AI requires businesses to rethink their data infrastructures with an eye toward unifying data regardless of where it is located.

“If you look at the applications that clients are trying to drive, whether it’s advanced analytics or AI applications of various types, the organizations that are further ahead are the ones that have unified their data, that have catalogued it well, that have a good understanding of security and controls for sensitive data, and that have a good handle on how the data changes over its life cycle,” he said. “Interestingly, regulated industries, for example in life sciences, where the responsibility to have good, clean, high-quality data that’s catalogued with the right clinical data codes or treatment codes, enables them to actually apply AI techniques a lot quicker than unregulated industries that really don’t have unified data.”

To learn more about how AI and intelligent data storage have become so intertwined, and about what it might take to make data work in an AI world, here is what Kurian had to say about the past, present and future of data and intelligence.

The Era Of Data And Intelligence

Kurian said the IT industry is at a really interesting point in the evolution of the business landscape and the technology landscape that he called the era of data and intelligence. This, he said, results from three intersecting trends that have reached the point of crescendo: human interest in understanding human behavior and then getting from understanding to insight and prediction, the need to collect data to understand human behavior, and the availability of mathematical tools used to normalize and standardize the data to derive insight.

“You can even see in the modern deep learning tools and machine learning tools the use of mathematical techniques that were created in the 1800s and 1700s by [Carl Friedrich] Gauss and [Pierre Simon Marquis de] Laplace and [Adrien-Marie] Legendre and various other types of mathematicians,” he said. “So we’ve had this long journey to get here, but where we are today is have much larger data sets, much more powerful tools, and therefore the responsibility to use them wisely to improve the human condition.”

The Third Phase Of The Modern Data Revolution

From a data perspective, the industry is in what Kurian called the third phase of the modern data revolution. The first phase, he said, was digitizing data, which has happened over the last 30 to 40 years. The second was starting to piece together the longitudinal history of a particular type of data to do trending and analysis.

“Now is the need to go wide in scope of the data, in addition to being long in trend, so that you can fully understand all of the data in your organization so that you can get the best results,” he said. “So for example, you want to unify the transactional data about your customer together with your most recent videoconference and your customer call center record all into one unified view of your customer. This need to unify all of your data continues to grow in importance.”

Four Keys To Success In The Era Of Data And Intelligence

Kurian said there are four keys to success in the era of data and intelligence:

Have a good data strategy and organization
Have powerful tools to analyze the data and the deep domain knowledge and insight about that data to know what you need for your business and understand the conclusions of various types of analysis
Have the ability to test, learn and adapt quickly
Have a data ecosystem, in addition to a business ecosystem, that provides a rich view of your data

“NetApp will keep helping our clients [stay] focused on the first part of that journey, which is their data strategy and how you build an infrastructure that enables you to manage your data and use it for insight,” he said.

Challenges Organizations Face With Their Data

Businesses are facing three key challenges when it comes to their data, Kurian said.

The first is the data management challenge they face as they move from siloed data to unified data, he said.

“If you look at the applications that clients are trying to drive, whether it’s advanced analytics or AI applications of various types, the organizations that are further ahead are the ones that have unified their data, that have catalogued it well, that have a good understanding of security and controls for sensitive data, and that have a good handle on how the data changes over its life cycle,” he said. “Interestingly, regulated industries, for example in life sciences where the responsibility to have good, clean, high-quality data that’s catalogued with the right clinical data codes or treatment codes, enables them to actually apply AI techniques a lot quicker than unregulated industries that really don’t have unified data.”

The second is a gap between AI systems and data systems, Kurian said, where AI systems are often being built in a silo with specialized chips and specialized networking, but they haven’t access to data, he said.

“Without data, the systems are useless,” he said. “We see this much like what cloud was in the early eras, where cloud was a distinct silo, the on-premises world was a distinct silo, and you needed to bridge the two. And so what NetApp is doing is much like building a data fabric to unify on-premises and cloud data and make it easier for clients to operate in a multi-cloud environment. We are actually solving the problems for the era of data intelligence by bringing AI to your data.”

NetApp is doing so with three innovations, Kurian said:

Bringing intelligence to the infrastructure and to the data that sits on the infrastructure with tools that businesses use to understand and explore where all their data assets are and quickly select the data needed for AI experiments
Bringing AI to customers’ infrastructure in a way that’s agile, attainable and secure so they can apply AI capabilities to all their data wherever it exists on-premises or in the cloud by adding things like traceability, data versioning and the ability to move data from the infrastructure into the applications that support AI like vector databases
Making it easier to maintain the security, privacy and control of data over its life cycle to help businesses track which data has changed and ensure all of the controls and security policies that have been implemented over the life cycle stay with the data

“I’m super excited about our advancements there,” he said. “You will see the work that we’ve done over the last many years with the cloud providers bring enormous value in the AI journey as well because those cloud providers are increasingly building AI and data platforms.”

How AI Will Change The Future Of Data

The idea of using machines to analyze data has made continuous progress over the years, and the last five to 10 years has seen functional improvements along two dimensions, Kurian said.

The first is the software algorithms that form the underpinnings of large language models have proven to have a deep humanlike ability to understand the domain in which they’re operating without human assistance, and the ability to switch from one domain to another, especially with the use of multimodal technologies like voice, video and text, he said. The second is the applicability of those algorithms to address a huge amount of data that enterprises around the world have.

“About 90 percent of the data in a large enterprise is unstructured data, and until now there was no systematic way to understand [it], organize it and draw insight from it,” he said. “Where we see the revolutionary applications today are, for example, in life sciences, where the fact that their data sets were well organized gives them the ability to find profoundly new capabilities. We have examples of new drugs that were discovered out of existing data sets because of the capabilities of these AI tools. … Broadly speaking, we see today applications for productivity improvements, better customer experience, better understanding of customers, and we see those progressing from those kinds of back-office and productivity applications as people’s data gets more mature into these transformative use cases.”

The Impact Of Regulation On Data

NetApp recognizes that there are risks inherent in AI, just like with any powerful tool, Kurian said.

“The idea of aggregating data to draw insight naturally can lead to concerns about bias and improper use,” he said.

But there is plenty of room for optimism, Kurian said.

“Our view is that if regulation is risk-focused, if it is particularly protecting the most vulnerable people like consumers that don’t have the ability to advocate on their on their behalf, that if it is adaptive, meaning that it recognizes the advancements in technology and then supports the advancements of technology to mitigate risk, and if it’s balanced, together with industry self-regulation and consumers being able to have agency on their behalf, that will help the overall world use AI more confidently,” he said. “And I think that will be a good outcome for everyone.”