The Economic Times daily newspaper is available online now.

    India's ChatGPT boom faces a reality check as free AI use meets rising costs

    Synopsis

    India has emerged as ChatGPT's largest market, driven by its vast digitally connected population and widespread access to free AI services. However, as AI infrastructure expands and computing and energy costs rise, companies face the challenge of recovering investments. Convincing price-sensitive Indian users to pay for AI remains difficult.

    ChatGPT running on a smartphone in front of a laptop. Photo: Jernej Furman (CC BY 2.0)Agencies

    India's rapid adoption of free AI services may prove difficult to monetise as infrastructure costs rise and users remain reluctant to pay for advanced capabilities whose productivity benefits are yet to be firmly established. (Representative image)


    India has become the largest market for ChatGPT. This follows a familiar script of technology adoption that makes it a global node for AI dispersal. Few markets compare with India's sheer population of digitally connected users.

    Data is vital to the development of AI models. Critically, though, the country also has one of the world's largest pools of trained manpower that can deliver secondary innovation with emerging technology. This gives India a rare optimism despite AI's disruptive potential.

    Low levels of AI infra development in India have muted opposition to the mounting resource needs of data centres. Of course, the most compelling reason for the rapid rise of AI use in India is that it is free for a vast majority.


    Consumer AI works on a freemium model where basic versions are made free and advanced capabilities require a paid subscription. AI hyperscalers use this to build user bases. Investments in AI infrastructure subsidise free user accounts and the interaction data helps improve underlying models. But the productivity gains from AI are expected to emerge from the paid tier of services.

    Only a small percentage of users pays for advanced queries, and this segment is now exposed to sticker shock. Companies are realising higher model consumption does not always translate into a boost in employee productivity or cost savings. Operational expenses of AI coding have, in some instances, exceeded human coding.

    Indian AI usage is a 'free rider' problem. People are benefiting from AI services without contributing to their cost. This changes when sovereign AI infra begins to take shape. The energy and computing costs will have to be recovered from consumers.

    That is a tricky proposition in India's price-sensitive market. A large part of the country's digital infra is built with public money, and the question of user fees remains a sticking point well after universal adoption of digital payments. AI has yet to conclusively establish its benefits for more price-sensitive (read: parsimonious) Indian users to open their wallets.

    Add ET Logo as a Reliable and Trusted News Source

    (Catch all the Business News, Breaking News, and Latest News Updates on The Economic Times.)

    Subscribe to The Economic Times Prime and read the ET ePaper online.

    ...more
    The Economic Times

    Stories you might be interested in