“AI sprawl is becoming the biggest enterprise risk now”

Enterprise AI is entering a new phase.

- [e-mail]

What began as isolated experiments with chatbots and copilots is rapidly evolving into large-scale deployments of AI agents across customer service, operations, software development, compliance and business processes. As organisations race to embed AI into everyday workflows, they are discovering that scaling intelligence is not simply a technology challenge. It is increasingly becoming a governance challenge.

Multiple teams are now building agents across different frameworks, cloud platforms and model providers, often at a pace that outstrips enterprise oversight mechanisms. The resulting complexity is giving rise to what many industry observers describe as “AI sprawl”, a fragmented environment where visibility, accountability and control become difficult to maintain.

Prasanna Arikala, CTO and Head of Products at Kore.ai, believes this challenge will define the next phase of enterprise AI adoption. In this interaction with Shubhendu Parth, he explains why governance must evolve alongside innovation, how enterprises can establish control across heterogeneous AI environments, and why a central operational layer is becoming essential for managing increasingly complex agent ecosystems. Excerpts:

As enterprises deploy multiple AI agents across platforms and clouds, how significant is the challenge of AI sprawl, and what risks does it introduce?

I think AI sprawl is already one of the biggest enterprise AI challenges, and it is happening much faster than most organisations anticipated.