Machines are now able to learn from experience and perform tasks that, not so long ago, were only possible `for humans. Technologies that make this possible are commonly called artificial intelligence or AI.

AI’s potential is now becoming clear. Unlike the automation of tasks that technology first made possible AI is proving revolutionary. Already AI is working away in the background in many industries, dealing with customers, identifying problems before they happen and saving time and resources. AI is also challenging company culture as humans and machines must now work together seamlessly in teams.

Ramesh Shanmuganathan, the Chief Executive of JKIT and the Executive Vice President and CIO at John Keells Holdings (JKH) Sri Lanka’s largest company, has been designing and deploying AI in several industries. Shanmuganathan is a leading innovation advocate both in the businesses he oversees and elsewhere. In an interview, he discussed how companies can deploy artificial intelligence to boost productivity and spark innovation.

Innovation is crucial for an organisation to succeed. How has the innovation agenda and mandate evolved?

Innovation is key to an organisation today because the world we live in is so dynamic, complex, integrated, volatile, and the list can go on. To sustain and grow, organisations must be agile and nimble – must be able to respond to stimuli, internal or external, and that too within a finite window that is shrinking with time.

Traditionally, innovation was vested with a unit within the organisation with the expectation it will drive advancement through research and development to drive that periodic step change in product or services they offer to create the points of inflexion. Today, most are realising, that alone, will not advance their growth or the innovation agenda. This has brought about the stark reality that for an organization to sustain and grow, the entire organisation needs to be equipped with the tools to drive the innovation agenda across the board.

This means companies must let creativity permeate their corporate culture, encouraging employees to seek creative solutions to challenges they face.

To make innovation a competitive advantage, we need for it to evolve beyond being an activity a few are engaged in and become a state of mind across the entire organisation.

What is important to drive this renewed mandate for innovation? Is AI a key differentiator?

The innovation mandate has changed because the context in which we operate has changed due to accelerated tech adaption and connectedness amongst constituents, be it – countries, organizations, customers, partners, suppliers or people.

We have also seen data becoming a core piece to drive innovation. We know that raw data is of not much use to drive anything, let alone innovation. Data needs to be correlated to give us meaningful insights to be able to leverage the same in either creating an opportunity or solving a problem.

In addition to availability and access to data, the advancement in high-performance computing and its wider adaption with increasing affordability and the progress we have made in cognitive & neural sciences have helped us to accelerate data-led innovation and bring it to the mainstream during the past decade.

Given the factors that I spoke of, data and technology can be acquired by various means but like a brain is to a human being, in a data-driven organization the AI is the key differentiator which creates the intelligence to drive value creation. That’s why AI becomes the game changer!

AI has evolved, what’s the current state of play? How does an organisation decide as to what they could do with it?

Yes, AI has evolved and can be broadly classified into two based on functionality and technology. It’s good to have a clear understanding of these so one can moderate expectations on what can and cannot be done.

AI platforms based on functionality:

• Reactive Machines – They emulate the human mind’s ability to respond to different kinds of stimuli. These machines do not have memory-based functionality.
• Limited Memory – in addition to having the capabilities of purely reactive machines, they are also capable of learning from historical data to make decisions. Nearly all existing applications that we know of come under this category of AI.
• Theory of Mind – A theory of mind level AI will be able to better understand the entities it is interacting with by discerning their needs, emotions, beliefs, and thought processes
• Self-Aware – AI has evolved to be so akin to the human brain that it has developed selfawareness.

AI platforms based on technology:

• Artificial Narrow Intelligence (ANI) – AI systems that can only perform a specific task autonomously using human-like capabilities. These machines can do nothing more than what they are programmed to do, and thus have a very limited or narrow range of competencies.
• Artificial General Intelligence (AGI) – AI agent to learn, perceive, understand, and function like a human being. These systems will be able to independently build multiple competencies and form connections and generalizations across domains, reducing on time needed for training.
• Artificial Super Intelligence (ASI) – become by far the most capable form of intelligence on earth. ASI, in addition to replicating the multi-faceted intelligence of human beings, will be better at everything they do because of greater memory, faster data processing and analysis, and decision-making capabilities.

Though AI aspirations cover the above only a few of those such as Reactive Machines, Limited Memory, as well as ANI, are in deployable readiness today. The rest remains either at a conceptual or research stage.

Ultimately any business leveraging AI for innovation would simply be making the most of its data – the right business use cases, right data sets, right talent, right models/analytics, right business models and process to execute it end-to-end.

How does an organization figure if AI-led innovation is for them? What should they look at in assessing whether they are ready for it?

I would reckon one has to broadly assess five areas to determine the readiness of their organizations for embracing AI-driven innovations,

• Strategy
• Structure
• Systems
• Skills
• Staff

And pertinent questions to answer for themselves in exploring the above would be,

• Is it an integrated roadmap across the company or targeted at point solutions?
• Do we executive sponsorship from top-down and across the organizations?
• Do we have a prioritised set of use cases with the respective business cases identified?
• Do we have access to requisite partnerships and industry expertise to mobilise?
• Have we assessed the impact on the business and revenue model to facilitate an end-to-end execution?
• Have we looked at privacy, security and regulatory concerns and restrictions in assessing the business case?
• Do we have clarity on how each business case will be executed on the ground with measurable outcomes?
• Do we have the nucleus in terms of AI Center of Excellence to coordinate and drive the same?
• Do we have access to a cross-functional team of experts from the intersection of business and technology to realize the above?
• Do we have a clear data governance strategy and structure to unify data across the siloes and manage the same for its value potential?
• Do we have the appetite and a business model to experiment with Data and AI?
• Are we investing in skills that can challenge the norm and conventional wisdom in being able to leverage data and AI progressively?

The above isn’t an exhaustive list, but it broadly covers areas that would be key to any data-driven initiative or AI adoption. The key is to take a holistic view rather than a siloed view.

Forward-thinking organizations globally, have integrated AI into their core business activities. AI has become an important part of business strategy. How ready are Sri Lankan businesses to embrace this wave? What must Sri Lanka do to accelerate adoption?

As you can appreciate by now, the journey to becoming a data-driven organization needs a certain level of maturity. It’s not a switch that you can flip readily. It needs a medium to a longer-term strategy for business, technology, data, culture, talent, and skills. The more forward-thinking organization have mostly been in step with these to be able to easily embrace the same whilst the rest will have to leap-frog with significant investments.

Most Sri Lankan companies, are, unfortunately, at the latter stage and have to make that quantum leap to do anything meaningful in this sphere. But, the good news is that with this becoming mainstream it’s becoming easily accessible through partnerships via readily available platforms and ecosystems that businesses can leverage on an, as a service model (XaaS), which solves mainly the technology aspects of the challenge. John Keells IT provides an end to end portfolio of solutions, platforms and services in this respect.

This still leaves the question of business use case, requisite data and the talent and skills to piece these together to make a working prototype that could be tested and refined to yield the required outcome. This is where we face our biggest challenge as a country – talent and skills. We must address the shortage from the supply side of the equation whilst attracting investments into the country which could accelerate the same and boost the confidence of local companies to take that leap of faith in making those investments.

The key pitfall to avoid is to view AI as a plug-and-play technology with immediate returns

For AI to function and yield results data is core, which for many organisations can be a challenge as data remains fragmented. How can organisations overcome this challenge?

Yes, data is the most important ingredient. We need to consider the attributes needed to validate data for it to be useful, and they can be classified into several key areas.

• Variety – data from structured sources to unstructured sources
• Volume – quantum of data from each data source
• Velocity – how fast the data changes from each source
• Veracity – the accuracy and quality of data from each source
• Value – the perceived value of data from each source

Once an organization is clear on what it wants to achieve with innovation, it becomes imperative to understand the holistic strategy for it and then to acquire data and build a data governance model for it, incorporating the requisite data sources and based on the above attributes. Then one has to deploy the requisite platforms and tools to manage the same.

Can AI improve outcomes in organisations, and how can one quantify its impact on an organisation?

Anything done well will yield value, and AI is no exception, but there are a lot of dependencies as discussed above and we need to take cognizance of these facts. The benefits or returns from AI can be broadly classified into three categories, Operations based ROI – Operational outcomes such as financial and quantitative returns such as high revenue, lower costs, better margins, better term value, etc.

Strategy based ROI – Strategic outcomes such as being able to create better brand value with better customer engagement, experience, lifestyle changes, disruptions/ shifts in markets, etc.

Maturity based ROI – this is the organization’s AI maturity in being able to leverage the capability beyond operations to change the game and being able to create that point of inflexion.

Most organisations would start the initiatives on an operational level and then progress to strategic and then finally graduate to the maturity level. It requires significant investment.

Can you talk about your personal experience of advancing AI linked innovation at John Keells? What would your advice be for others who may be planning the same?

Our journey commenced in 2015 with many exploratory interventions. But these were siloed around use cases at both John Keells as well as with external customers such as fraud analytics, basket analysis, employee/customer churn prediction, predicting flight arrival/departure times, customer profiling & engagement, etc as John Keells IT.

At John Keells, we moved to the next phase of maturity with a very structured program addressing many of what I had discussed above in a holistic sense. We now have a welldefined program, coordinated top-down. As we navigate the trajectory across the group we have identified potential use cases which can benefit from this intervention and brought them into the mainstream bringing the constituents into a well-planned and sustainable program. However, we are at an early stage of this journey, given the scale of opportunities we can leverage across the group.

The key pitfall to avoid is to view AI as a plug-and-play technology with immediate returns. One needs to balance investments vs return whilst scaling from a few pilots to company-wide programs whilst the focus shifts from discrete business problems, such as improved customer segmentation, to big business challenges, like optimizing the entire customer journey.

It’s also important to think broadly about the entire AI strategy and journey in assessing technology, talent as well as to equally align the company’s culture, structure, and ways of working to support broader AI adoption. Bear in mind that businesses aren’t born digital. Traditional mindsets and ways of working run counter to those needed for AI.

Let the data be the lighthouse in the new normal, and AI be the tool that helps navigate the ship.