Artificial intelligence (AI) is finally coming of age in India. We already have a large, technology-savvy young population working in various fields of digital technology, both in services and products. According to industry body Nasscom, over 1.5 million engineers graduate annually. The IT industry alone employs almost 5.5 million engineers directly. India is the second-largest generator of digital data, only behind China. It is the fifth-largest economy in the world today and one of the biggest markets for any technology, both in business and customer applications. The country ranks among the biggest markets for almost all big tech firms. The Union government’s Digital India initiative and the robust IndiaAI Mission plan, combined with the impetus to semiconductor manufacturing through the PLI scheme, has put in place the key building blocks.
The interest shown by the private sector has also contributed to this momentum. Big companies have been experimenting with AI through pilot projects and limited deployment for several years. With the advent of Generative AI (Gen AI) and its visible benefits, Indian companies are recommitting themselves to scaling to deliver business value and gain a competitive advantage. While some are building an AI strategy from the ground up, others are taking one use case or application to improve efficiency, manage risks, or improve experiences in their core areas.
For example, a leading Indian non-banking financial company has built a Gen AI agent to put together and analyse a potential borrower’s balance sheet and profit and loss (P&L) documents, along with business information available about the company as well as the sector on the internet, to generate a credit assessment memo. Underwriters can, therefore, focus on decision-making rather than gathering information on SMEs. A big Indian retailer is using a customised Gen AI agent to offer a far more personalised experience to its customers than its older generation of rule-based AI could. Another large retailer is leveraging AI to reduce employee churn on the shopfloor.
As a professional services firm, we, too, have been increasingly using AI to provide our clients with a better experience. For example, AI integrated into our vast tax knowledge repository allows us to offer more efficient and faster analysis for clients.
What CEOs must consider before accelerating AI strategies: Corporations undergoing digital transformation for the past few years are now leveraging AI alongside digital, cloud, and automation capabilities. Those who were moving slowly are accelerating. Four ingredients are particularly critical in this effort. The first is identifying areas where the technology will deliver maximum impact and the highest ROI (return on investment). CEOs must assess reliability and do a rigorous cost-benefit analysis.
As we argued in our recent report, “AIdea of India: Generative AI’s potential to accelerate India’s digital transformation,” the cost implications of full-scale AI deployment are huge and leave little margin for error, as there are significant one-time investments and recurring operational costs. CEOs and senior leadership should roll up their sleeves and try out various AI tools to understand the real challenges and opportunities—it wouldn’t be a surprise if they uncover a new course for the organisation and the entire industry!
Also, there are far too many flavours of AI available today, in both open-source and closed-source AI models. Organisations also need to compare customising ready-made solutions versus building their own.
The second factor to consider is that large-scale AI implementation is not just a case of overlaying the technology on top of the existing IT systems; in order to get the best results, the firm may need to remodel the current technology infrastructure and processes. In some cases, work processes may have to be re-engineered with an AI-first outlook. Legacy data as well as data being generated and captured need special attention. While most big AI foundational models can use both structured and unstructured data to train from, organisations need to ensure that the data is clean, accurate, and above all, without bias before starting to use the data for AI training.
The third aspect is to have in place a proper AI governance structure. Even as policymakers globally develop regulations, companies, too, must build guardrails to prevent misuse by employees or outsiders. Ethical and responsible AI also requires checking for biases that can creep into either the data or the weights used in the algorithms.
When it comes to the human capital dimension, one could argue that there are two patterns playing out. First, there are mixed signals in some quarters with respect to the impact on jobs. A balanced and pragmatic approach, combining the best of humans and technology, would not only be valuable but also the most sustainable strategy. Simultaneously, Gen AI is expected to create and exacerbate talent shortages as more organisations embark on their transformation journey. The best way to address the shortage and deal with the impact on jobs is to retrain the workforce in AI skill sets. Enterprises may need to recruit individuals with natural language processing (NLP) and traditional AI/ML backgrounds, followed by providing them with targeted training.
The productivity benefits of a wide-scale AI deployment in the private sector combined with new AI-based business opportunities and ecosystems can truly supercharge the Indian economy, adding trillions to the country’s GDP annually over the next few decades. The government and the private sector must continue working together to take full advantage of the AI opportunities. After all, AI holds the promise of changing lives and the economy in the same way that electricity did two centuries ago. 
Rajiv Memani is Chairman and CEO of EY India. Views are personal