EY NASSCOM AI Adoption Index: is AI still incubating in your organization or driving innovation?

The EY NASSCOM AI Adoption Index, reveals that India is leapfrogging in AI maturity, but some dichotomies still exist.

In brief 

• The EY NASSCOM AI Maturity Index aims to measure and analyze the preparedness of Indian enterprises in converting the AI opportunity into tangible economic value, year-on-year.
• As per the index, ~60% of AI-led value-add is expected from five sectors.

Investment: A majority of the organizations in the intermediate stage of AI maturity have need-based or basic AI budget allocations leading to lower percentages of IT spend on AI. It still is a shift from 2020 when vast majority of corporations did not have an AI budget.

Data and technology: 1 out of 2 organizations surveyed have standardization and data readiness at either BU or enterprise level. A majority of organizations interviewed also indicated a shift towards an enterprise wide data lake in the near future to improve data readiness and access.

Knowledge output: Organizations prefer AI start-up incubation and in-house AI-labs for innovation. However, there is a limited focus on sponsoring AI research. Expanded industry-academia collaboration can bring diverse points of view and lead to innovative solutions.

Ethics, governance and controls: 3 out of 5 organizations have duly documented ethical and governance framework in place. With increasing role of AI applications, the need for an integrated AI ethical framework becomes very important.

A look at the AI footprint in core sectors

As per the overall sector scores, all the sectors fall in the “Enthusiast” stage. This signifies that all the sectors are beyond their phase of exploring the possibilities of AI and are now enthusiastic about what they can achieve by leveraging AI.

Industrials and Automotive: Sector Score: 2.52 | Level:  Enthusiast

Industrials and Automotive sector has traditionally lagged in its AI adoption pace but is gaining momentum with focus on Proof of concept to-production, formal AI strategy, and Cloud investments. The organizations in this sector use AI majorly to improve shopfloor operations with 24x7 operations, lesser defects and lower downtime.

Automotive players additionally use AI to drive innovation and product/service development in form of autonomous, self-driving vehicles, and aftermarket services like predictive maintenance and insurance etc.

Banking, Financial Services and Insurance (BFSI) : Sector Score: 2.42 | Level:  Enthusiast

30% of the surveyed BFSI companies, across size and sub-sector classification, are yet to begin a planned AI journey and are still exploring AI for applicability, viable use cases through PoCs, and a clearly explainable RoI.

The sector has been indulging in producing novel solutions, such as digital lending, fraud detection and auto underwriting, to improve customer experience and reduce operational costs using AI. The momentum is now shifting towards differentiation and ensuring data safety via AI.

CPG and Retail: Sector Score: 2.51 | Level:  Enthusiast

The industry is in a perpetual state of upheaval and development, contending with rapidly changing customer buying patterns and a shift in focus from the street to the online. There is a constant need to build novel solutions to improve customer experience catering to their ever-changing need, with a huge emphasis on consumer data privacy continuing as the key area of focus for the future.

The sector has expansion, optimization and innovation at the core of its AI initiatives with undefined or functional level AI strategy. The focus is now on scaling-up the ongoing efforts, developing a structured PM methodology, transitioning to modern systems and data standardization.

Healthcare: Sector Score: 2.35 | Level:  Enthusiast

47% of the surveyed healthcare companies, across size and sub-sector classification, are yet to begin a planned AI journey and are still exploring AI for applicability, viable use cases through PoCs, and a clearly explainable RoI.

For a data-intensive sector that can benefit immensely by a shift to AI-led preventive healthcare strategies, from predominantly curative, this indicates a massive need for AI advocacy and RoI demonstration.

Impediments to AI Adoption

The impediments to adoption of AI vary greatly from one sector to another. However, an aggregate view demonstrates that objectively quantifying the benefits of AI remains the top barrier to adoption and there has been no change since 2020. This is one area where there has been a gradual shift to Proof-of-Value in addition to Proof-of –Concept to demonstrate organizational value-add.  Similarly, the external enabler ecosystem has remained passive as indicated by the majority of the respondents across verticals. Having a budget for AI projects is the chief concern for the Industrial and Automotive sector as well as Healthcare. Thus, the top 3 challenges remain unchanged since 2020. However, cultural and behavioral impediments have moved within the top 4 in 2022.

Survey findings indicate a strong desire and the know-how to adopt AI, and a reasonable sense of the roadblocks ahead. A majority of enterprises seek to resolve their operational efficiency and market growth challenges with the use of AI. Many of these objectives can be realized with already proven AI use cases – either adopted by more progressive enterprises in India or by global companies. What is needed to quickly implement those solutions is a decisive leadership vision and AI strategy, strong technology infrastructure and data standards, and clear outcome definitions. 

India has world-leading AI talent. However, it lacks in skills that need a combination of depth of AI experience in specific domain areas to develop, test, and deploy focused AI solutions. This challenge further exacerbates when combined with AI-specific project management experience, an area where Indian enterprises seem to be lagging significantly. 

Indian enterprises prefer AI start-up incubation or inhouse AI labs to drive innovation, however, there is limited emphasis on partnership with academia to quickly create concepts that have a high potential for lab-to-market success. A majority of the patent filing within the country is driven by research and academic institutes. However, very few get implemented as real-world AI solutions.

While companies have started experimenting with AI, the frameworks to measure success continue to be legacy – project based RoI or time and budget success. However, digital transformation projects, increasingly with most having AI solutions embedded in them, require a different set of metrics to be defined and linked with organizational success KPIs, to measure the continual and cross-functional impact of AI.