The enterprise AI payoff shifts beyond models to mission-critical workflows

Enterprise AI capabilities are improving, but returns are still lagging behind spending. The technology is moving into production, yet it often fails to reach the business processes where revenue, innovation, and risk reside. This gap is particularly pronounced in industries where a wrong answer can have significant consequences, such as pharmaceutical research, financial services, and national security.

Most of the adoption in recent years has been focused on individual desktop productivity rather than core operations, with 57% of organizations still struggling to generate returns that exceed their investment, according to Thomas Robinson, CEO of Domino Data Lab Inc. Robinson asserts that many of the applications seen over the past few years have been about putting tools in the end-user compute context, such as drafting emails or creating marketing copy, rather than in high-consequence, mission-critical business processes.

The issue lies in measurement, as counting seats, tokens, or raw adoption indicates spending levels rather than generated revenue. Framing projects as solely cost-reduction initiatives limits potential upside before the work even starts. Leaders who focus on AI's potential for future innovation, new product development, and revenue generation are more successful.

Trust is another constraint, especially as agentic systems scale ahead of proper governance. Approximately 41% of organizations are either piloting or scaling agentic AI without the necessary oversight. Domino Data Lab addresses this gap with a three-layer model: a policy engine to gate projects from initiation through delivery, continuous monitoring and tracing in production, and a human accountable for the outcome.

Robinson emphasizes that while models are being hailed as reasoning tools, he believes human judgment remains superior. This is why he expects enterprise AI value to shift away from models and toward integration, governance, and workflow. Companies that build forward-deployed engineering teams to implement these solutions are driving digital transformation and capitalizing on the new technology.