Shadow AI: The emerging enterprise risk that can no longer be ignored

By Praveen Ojha, Chief Technologist, EPAM Systems India Pvt. Ltd.

As GenAI (Generative AI) accelerates across enterprises, a second parallel ecosystem has quietly taken shape: Shadow AI. Much like shadow IT in the cloud era, these are AI tools, models and workflows adopted by employees outside of formal governance. Their intent is often efficiency, but their impact is disproportionately risky. What makes Shadow AI more complex than its predecessors is the speed of adoption, the invisibility of model interactions and the sensitivity of the data now flowing into external systems.

Recent research from MIT highlights this gap: employees at more than 90% of surveyed companies are using personal AI accounts for daily work tasks, while only 40% of organisations provide official LLM (large language model) tools. This data illustrates the core challenge: AI has become essential to productivity, but enterprise controls have not kept pace.

The Risk ...