A Pragmatic Approach to Adopting AI | Straighttalk


The latest insights from your peers on the latest in Enterprise IT, straight to your inbox.

Q&A with Kalyan Kumar, Executive Vice President & CTO - IT Services, HCL Technologies

Artificial intelligence, and the related concept of automation, have moved from discussion topic for CTOs and CIOs to agenda item in the board-room. Investments are flowing rapidly into the development of AI technologies—including deep learning, machine learning, natural language processing, cognitive and computer vision—both at the startup level as well as within some of the world's leading technology companies.

In this interview with Kalyan Kumar, CTO of HCL Technologies, he offers some guiding principles for companies as they adopt AI technologies.

How can enterprises begin to implement and reap the benefits of AI?

You start with automation, which allows companies to improve efficiency and reduce costs by streamlining repetitive processes. But that’s just the beginning. You need to combine automation with artificial intelligence/analytics, then orchestrate the application of those technologies in an interconnected manner to business tasks across infrastructure, applications, business processes, and product engineering. This AI-enabled automation, or autonomics, can “set an enterprise free,” transforming it into a true 21st-century enterprise, a lean operation that is as agile as a start-up.

Why is orchestration so important?

The intelligent automation of individual business task has benefits that are limited by the nature of that task. But when autonomics is applied to a task in concert with other tasks—when that task becomes part of a streamlined workflow of tasks—the results are powerful. You end up with a connected organization in which different functions work collaboratively across traditional silos, using stateless digital applications, and an increasingly software-defined infrastructure. Through orchestration, you end up with a lean and agile enterprise that can quickly respond to opportunities—one that moves beyond simply greater efficiency and becomes truly alive, able to respond quickly to changing market needs.

[can this para for space, if necessary] Achieving this requires an “orchestrator” whose vision is broad enough to see the relationships between those tasks. In music, orchestration involves the writing and the arrangement of music for an ensemble. The orchestrator assigns different instruments and different musical parts to individual musicians. The onus of producing the desired overall musical effect is on the orchestrator, not on the individual performers. In the context of IT Services, orchestration involves coordinating and directing IT operations in an intelligent and holistic manner. And, as in music, the onus of delivering the desired overall business outcomes lies with the orchestrator.

How should companies go about deploying autonomics to ensure a return on their investment?

The approach that will generate the greatest business value will be pragmatic rather than rigidly prescriptive. It will leverage both proprietary and open sources technologies. It will draw on a collaborative ecosystem of expertise rather on than a single source. And it will involve a careful assessment of both an enterprise’s current IT system maturity and its business goals. This assessment will allow for the segmentation of complex and mixed IT environments into “zones” that permit the creation of a practical IT autonomics roadmap, one that will help IT to control spending and set realistic automation goals for its business stakeholders.

These principles hold true whether you’re looking at a standard environment, in which a solution can be essentially plug-n-play, or a complex legacy environment, in which the solution will need to be customized to integrate with the legacy structures.

What about the concern that AI will make human beings redundant?

Adopting AI isn’t about replacing people, it’s about augmenting human capabilities. Artificial intelligence technologies enable business transformation by doing the unglamorous work that humans are not so good at—for example, processing huge amounts of data quickly, efficiently, and accurately.  The relationship between humans and AI is actually mutually empowering. The question that IT and business leaders need to ask is: Are we using AI to replace people or are we using AI to make them work better, faster and more efficiently than ever before. In other words, are we using AI to divide or to multiply human potential.