Why Indian companies are struggling to scale AI despite rapid adoption: An expert explains

Indian companies are facing hurdles in scaling artificial intelligence (AI) despite its rapid adoption, due to issues with data quality, security, and a shortage of skilled professionals. Sriram Raghavan, General Manager of IBM Software for India and Software Innovation Labs, discussed these challenges in an interview with The Indian Express.

Enterprises are now grappling with ensuring reliable data availability to AI agents, managing identity and security risks, and maintaining regulatory compliance as they deploy the technology at scale.

Raghavan emphasized that while India should focus on building sovereign AI capabilities, achieving complete independence across the technology stack is currently impractical, mainly due to dependence on foreign hardware. He explained that both the government and enterprises care about control and choice, but the government looks at a much longer arc when considering sovereignty across all layers of the stack. However, achieving full stack sovereignty today is deemed impractical for enterprises.

Raghavan highlighted three key areas where enterprises can drive efficiency and control in AI deployments: power efficiency, platform optimization, and model architecture. While hardware prices rising can make the software layer even more crucial, immediate leverage lies in the software layer. IBM aims to assist enterprises in transitioning from experimentation to active deployment of AI by focusing on organizational transformation and technology.

They help companies identify architectural building blocks, create real-time context layers for AI agents, and address identity and security concerns.