Data modeling in the dawn of AI

In a rapidly evolving landscape where AI systems are increasingly mastering human-level tasks such as coding and drawing inferences, one pressing question emerges: Is data modeling still relevant?

Actually, yes. In fact, the demands of AI have made it more relevant than ever.

If this seems counterintuitive, think back to before the early 2000s when data modeling was a ubiquitous and inseparable part of database and data warehouse development. The reason was simple: the costs of not getting data quality and consistency right on the first go were staggeringly prohibitive due to limited storage and processing capacity.

Then came the internet and the era of big data, where two fundamental changes took place:

1. Due to the five V’s, relational systems could no longer meet all business demands
2. Technological advances such as cheap storage, distributed computing, and in-memory processing have lowered costs and data processing...