Enterprise AI Struggles With Context
A recent poll published by MIT Technology Review Insights indicates that only 16 percent of surveyed organizations are extracting transformative value from their artificial intelligence deployments. Despite widespread adoption of conversational assistants and workflow automation tools, companies frequently encounter confused or inaccurate outcomes because models lack a proper understanding of internal business operations.
According to the findings, the largest share of respondents at 40 percent report only moderate gains across specific tasks, while 23 percent see minimal returns or none at all. Most enterprises remain in the piloting phase or rely on basic employee assistance tools rather than full deployment.
Prukalpa Sankar, co-founder of Atlan, noted in the report that artificial intelligence models often possess high capability without being practically useful in specific organizational settings. The missing component is an architectural context layer that translates enterprise intelligence into governed, machine-usable data for autonomous agents and models. Without this foundational context, organizations struggle to bridge the gap between advanced technology and tangible business returns.
Based on reporting by www.technologyreview.com.
