Indian Firms Scale AI Without Replacing Legacy Systems

An Indian retailer's AI agent gets a simple job: process a customer order. Pull the customer's details from the CRM, check stock levels in the ERP system, trigger the next step. Three systems, one task. If any of them return conflicting records or simply won't talk to each other, the agent stalls, not because the model is bad, but because the plumbing underneath it was never built for this kind of work.

That's the problem Indian enterprises are running into as AI moves from pilot projects to daily operations. Dell Technologies' Modern Enterprise Readiness Study, done with Vanson Bourne, found that 67% of Indian organisations don't think their data centre environments can handle AI workloads at production scale. The models keep improving. The infrastructure underneath them, built up over decades of legacy applications and fragmented data systems, hasn't caught up.

Companies are responding not by ripping out old systems but by building around them, wrapping legacy architecture in new layers that let AI agents actually function across CRM platforms, ERP systems, and whatever else a business has accumulated over the years. What that looks like in practice, and why it's become the real bottleneck for enterprise AI in India, is where the harder questions start.