It has been two years since OpenAI released ChatGPT into the wild, racking up the fastest-ever 200 million users, setting off a feeding frenzy in the tech world, and heralding the inevitable Age of AI. As 2025 dawns and the technology takes its baby steps into some kind of maturity, I believe there will be four megatrends that will drive AI to shape business and society in India and across the world.

Apps to Agents: If 2023 was the year of ChatGPT, and 2024 was the year when multiple Large Language Models (LLMs) like Claude and Gemini bloomed into existence, 2025 will be the year of AI Agents, or Agentic AI. A prescient Bill Gates blog foresaw this in 2023. “In the next five years,” he wrote, “...you’ll simply tell your device, in everyday language, what you want to do...and software will be able to respond personally because it will have a rich understanding of your life. This type of software—something that responds to natural language and can accomplish many different tasks based on its knowledge of the user—is called an agent.” “Agents are not only going to change how everyone interacts with computers,” he went on to say, “they’re also going to upend the software industry.” Instead of tapping a few apps multiple times or asking ChatGPT a series of questions to work out a complex and frustrating travel itinerary, what if, say, a Booking.com agent would select a hotel and airline we usually prefer, design a day-to-day schedule based on our interests, and go ahead and book the tickets and hotels, after we have given it the permission and the agency to do so. Agentic AI will bring Generative AI to the enterprise, with AI bundling up discrete steps in a typical enterprise workflow into an AI agent software, replacing traditional Software-as-a-service into the new SaaS: Service-as-a-software. Bain Capital’s Sarah Hinkfuss elegantly describes it: “We are used to ‘pulling’ information from computers. (AI agents will) ‘push’ finished work to us instead.

Thousands of startups are amping up their innovation to build agents on top of the LLMs. Genie by Gather helps parents manage their time, Minday scours the internet and mines your preferences to find the best restaurant or shop around you, and Relevance AI automates prospect meetings for harried sales reps. Klarna, a European fintech, made waves when its CEO announced that customer service agents built on Open AI platforms have ‘replaced’ 700 human agents and that they resolve queries in a fifth of the time it took humans. The implications for India are profound. As contact centres deal with the human agent to AI agent transition, there is a vast new opportunity for Indian startups to create agents on top of global LLMs, and Indian enterprises will be well served to understand and incorporate AI agents in their workflows.

From LLMs to Applications: This brings the old question back—does India need its own LLM? Nandan Nilekani is adamantly opposed to it, saying, “Let the big boys of the Valley do it”, with India focusing on building use cases and applications on AI, rather than an LLM—a view I subscribe to. Much like the cloud, I think LLMs would be commoditised, with three or four major ones emerging as the infrastructure on which millions of apps and agents will be built. So far, most of the value in GenAI has been concentrated in the infrastructure layer, where Nvidia and the cloud providers rule; however, 2025 will see value migrate upwards to the application layer at the top which will distribute value to companies and startups innovating on top of these. This is the moment India should grasp, with its IT services firms and eager startups solving India-specific and global use cases across enterprises and consumers with AI agents and applications.

Becoming AI-literate: The definition of literacy will extend beyond reading, writing and arithmetic to using GenAI tools like ChatGPT, Perplexity and others to become more productive and efficient at work. Organisations are witnessing a new trend, BYOAI or Bring Your Own AI, where three of four employees are bringing along their own AI to work, because it helps them perform so much better (https://bit.ly/41GKJe4). Two-thirds of bosses will not hire an employee unless she has the aptitude and curiosity to work with the profusion of GenAI tools out there. Thus, it will be incumbent on managements across organisations, irrespective of sector or geography, to frame the right policies and build enablers for all their employees to become AI-literate. This is especially true in India, with a recent BCG report placing Indians among the top users of ChatGPT worldwide. We need to aspire to make every Indian AI-literate.