TLDR: Sriram Raghavan, Vice President at IBM Research for AI, emphasized that businesses and governments must embrace artificial intelligence as a comprehensive transformation initiative rather than a mere technological experiment. Speaking at the ‘Demystifying the AI Revolution’ seminar in Chennai, Raghavan highlighted AI’s increasing affordability and broad applicability, urging a strategic, top-down commitment to integrate AI deeply into core operations and address critical challenges like data disparities and responsible governance.
CHENNAI – Sriram Raghavan, Vice President at IBM Research for AI, delivered a compelling address at the ‘Demystifying the AI Revolution’ seminar, organized by the Chennai International Centre on July 20, 2025. Raghavan strongly advocated for a paradigm shift in how organizations perceive and implement artificial intelligence, urging them to treat it as a fundamental business transformation rather than a superficial technological add-on or a mere pilot project.
Raghavan underscored the critical need for a strategic approach, stating, “Don’t treat AI like a toy or a side pilot. You must approach it as a business transformation.” He observed that a significant number of AI deployments falter because organizations fail to integrate these systems meaningfully into their core strategies. He further elaborated on the nature of current AI systems, particularly generative models, noting that they do not possess a ‘world model.’ “AI doesn’t know what is right or wrong. It just knows what is likely,” he cautioned, stressing the importance of human oversight to prevent blind trust in AI-generated outputs.
Addressing the rapid evolution of AI, Raghavan highlighted its increasing accessibility and cost-effectiveness. He projected that the cost of performing tasks with AI would continue to decrease significantly, stating, “The amount of progress in making AI cheaper is enormous. You should assume that whatever task you are doing will get cheaper every six months for the next five plus years.” This trend, he believes, will ensure that AI eventually impacts nearly every domain, though the pace of adoption may vary across sectors.
On the crucial topic of regulation, Raghavan asserted that it is “not a question of if, but how,” emphasizing that India must develop its own governance frameworks tailored to its unique priorities and risks, rather than simply replicating models from the EU or US. He stressed the necessity of domain-specific guardrails, particularly in sensitive areas such as healthcare, finance, and education, to build trustworthy AI systems. He also flagged significant risks associated with AI, including issues of fairness, authorship, accountability in automated systems, and the potential for data gaps to exacerbate existing disparities, especially for underserved communities and those using low-resource languages.
Raghavan called upon the public sector to lead in AI adoption, focusing on deep domain insight and inclusive data-building efforts. He clarified that the true value of AI lies beyond ‘flashy chatbot projects,’ advocating for its meaningful application in critical sectors like climate, agriculture, and health. For students and professionals, he offered pragmatic advice: “Don’t chase trends. Prompt engineers will come and go.” Instead, he encouraged building deep expertise in a chosen domain—from law to healthcare—and then exploring how AI can augment that specialized knowledge, cautioning against becoming ‘extreme specialists’ and advocating for broad learning alongside depth.
Also Read:
- Indian CIOs Charting a Course for Scalable AI Integration Beyond Initial Enthusiasm
- Small and Medium-Sized Businesses Face Urgent Need to Embrace AI for Future Success
Raghavan also touched upon the impact of AI on employment, noting that unlike previous technologies that automated physical tasks, AI aims to substitute cognitive aspects, leading to faster adoption and presenting significant reskilling challenges. He dismissed the narrative of AI becoming super-intelligent and taking over the world, instead focusing on tangible risks like misinformation, deep fakes, and cybersecurity threats that regulations should address. He concluded by advising enterprises not to wait but to ‘jump into the AI game’ with a long-term perspective, recognizing data as a key differentiator for successful AI integration.
