MIT Technology Review is reporting that enterprise AI systems can generate recommendations and automate tasks, but without an understanding of how a business operates, even powerful models can miss the constraints that determine whether a decision is right. In an episode of its Business Lab podcast produced in partnership with Celonis, the publication said companies moving AI from experimentation into everyday operations face the challenge of giving the technology context to act usefully within complex business processes.
Devavrat Shah, chief scientist of enterprise AI at Celonis, told MIT Technology Review that this context is central to what he calls "decision intelligence." He said AI agents can support individual tasks within larger decision-making processes but need to understand the role each task plays and the business conditions surrounding it. "You can't just do that in a vacuum," Shah said. Without the right operational context, he said, agents risk overlooking important constraints, making costly mistakes, or consuming excessive computing resources.
MIT Technology Review said the solution begins with connecting AI to the processes and data enterprises already use. It said a contextual layer can help agents understand workflows, operate within defined guardrails, and produce recommendations that people can review and approve. Shah advised organizations to start with a specific business decision, identify its most difficult steps, and test whether AI can improve them, then measure gains in speed and value to determine whether an application is worth scaling.
The publication reported that this approach could change how employees spend their time, with automation of repetitive manual work freeing people to focus on more complex strategic decisions. It said realizing that benefit will require investment in workforce education so employees can understand AI outputs and their implications rather than simply handing decisions to agents.
Looking ahead, Shah told MIT Technology Review that opportunities extend from routine operational tasks to enterprise-wide planning in areas such as finance, workforce management, marketing, and sales. He said AI with the right context could help organizations evaluate options and make consequential decisions more quickly. "Now, maybe [resource planning] should not take that much time, and people are actually making the right choices quickly," Shah said.
MIT Technology Review noted that the episode was produced by its Insights custom content division in partnership with Celonis.
Full Article: Making the case for context-aware AI