Enterprises must transition from treating artificial intelligence (AI) as a separate experiment to embedding it into core business processes to generate measurable value at scale, stated Manas Mehra, Global CIO of Dabur, during the ETCIO Annual Conclave 2026. As companies shift from exploration to execution, technology leaders confirmed that the next phase of AI adoption hinges on data readiness, process integration, governance, and responsible deployment.
Mehra noted that many organizations remain in the experimentation or pilot phase, warning that a limited number of AI initiatives will successfully scale unless integrated into existing business workflows. “Do not look at AI as something separate. Look at it as something that is embedded into what you are already doing, in a simpler and more efficient way,” he advised. He emphasized that AI should not be the initial focus in every transformation initiative; instead, businesses should start with identifying their challenges and determining where AI can enhance efficiency, decision-making, or execution.
Ujjwal Mathur, President of India Business & Strategic Accounts at TCS, highlighted that enterprises are progressing toward a stage where AI will inherently be part of technology initiatives. He drew a parallel to the earlier transition that saw digital solutions becoming a standard layer in enterprise projects. “We are reaching a stage where we may not call it an AI project. It will be a project where AI comes by default,” Mathur remarked.
Mathur also underscored the importance of data governance in scaling AI, warning that inadequate data quality can lead to incorrect outcomes and hallucinations. He stressed the necessity for robust data management, metadata governance, and oversight before scaling AI deployments. Furthermore, he identified cloud technology as a critical enabler for AI, offering infrastructure accessibility, GPU scalability, rapid deployment, and reduced barriers.
Moreover, Mathur cautioned that cybersecurity must be integrated into AI architecture from the outset. As AI adoption grows, the associated risks, including model security, data leakage, and identity compromise, will also escalate. “Cybersecurity has to be part of the AI architecture,” he stated.
The discussion also emphasized that privacy, compliance, and ethical considerations are essential components of enterprise AI. Mathur advocated for explainability and transparency in AI systems, enabling businesses to understand the rationale behind AI utilization and its decision-making processes.
The fireside chat concluded with a perspective that India’s AI journey is now entering a more execution-driven phase. For enterprises, the primary challenge will be to progress from isolated pilot projects to integrated, governed, and secure AI systems capable of delivering long-term business value.
(With inputs from Sachi Srivastava).
