The growing role of synthetic data in overcoming AI challenges
By Dr. Radha Krishna B, Director of Customer Advisory at SAS India
Imagine a healthcare startup in India trying to build an AI model to predict patient outcomes. The ambition is clear, but the reality is daunting. Patient data is highly sensitive, governed by strict privacy laws like the Digital Personal Data Protection (DPDP) Act, and acquiring large, diverse datasets is both expensive and time-consuming. This is not an isolated challenge. Across industries, organizations are grappling with the same dilemma: how to innovate with AI when access to real-world data is limited.
The challenge isn’t limited to scarcity but also spans across quality, accessibility, timeliness, and trust. These constraints multiply the cost of experimentation and increase operational risk. Models trained on limited or skewed datasets struggle to generalize. Decision engines perform inconsistently across segments and time periods. Teams spend weeks waiting for data access approvals, which delays deployment and dilutes...
Copyright of this story solely belongs to www.expresscomputer.in. To see the full text click HERE
