Poor data quality is breaking AI ambitions: Anand Ramamoorthy, Director APAC Data Governance and Quality, Informatica, Salesforce

Anand Ramamoorthy, Director APAC Data Governance and Quality at Informatica

Low quality of data is the largest hindrance within the enterprise leap from pilot stage AI-systems to manufacturing prepared AI brokers, as per Anand Ramamoorthy, Director APAC Data Governance and Quality at Informatica from Salesforce in an unique dialog with businessline.

Following the corporate’s Data and AI Summit in Mumbai on June 11, Ramamoorthy stated many enterprises voiced concern in regards to the quality of their data with one firm flagging 73 per cent of its data as “unhealthy”.

Data governance

“Data quality is the largest obstacle to translating these brokers that you just’re constructing to make it production-ready. It’s crucial to have a data governance functionality. The issue is the guide effort slows issues down. So, though it’s crucial, they lose endurance pondering that they’re not getting the worth,” stated Ramamoorthy.

Regardless of this hurdle, Ramamoorthy identified that conventional data administration norms count on people to be within the loop of workflows or within the deciphering of data. Because of this the context when utilizing data must be set by a human that may perceive nuance relatively than and AI agent that is liable to hallucinate.

“Data itself is not at all times incorrect. It is generally ambiguous, relying on the utilization, the interpretation could be completely different. So, while you apply the brokers on high of that, it amplifies that downside. That may result in hallucination, the outcomes could be completely different,” he stated.

To handle these considerations, Ramamoorthy urged three key pillars on the earth of agentic AI: machine readable metadata with trusted context, AI-ready data merchandise and AI-assisted data stewards that take up the duty of offering governance insurance policies round possession, accountability, definitions, and so on.

“It’s about how can we unlock the worth of those data stewards who’re good at purposeful and business-oriented conversations. How can we instill that into the guide effort relatively than give attention to mundane, repeatable guide duties?” he stated.

Printed on June 16, 2026