As agentic artificial intelligence moves out of its experimentation phase, how should supply chain leaders be rethinking the way decisions are made? Insights from Fab Brasca, senior vice president of market strategy with Kinaxis.

As with any new technology, agentic AI is innovating at a rapid pace, Brasca says. It’s gaining the ability to execute tasks at scale. He cites “the pure computational power available now, to allow agents to act on your behalf. You don’t have to wait a long time for the task to be done.”

Agentic AI is especially proving itself in its ability to engage in multi-scenario planning, “en masse and very quickly. It helps with the whole notion of efficiency, scalability and productivity.”

AI “co-pilots” are designed to augment human activities, but true agentic AI focuses on “autonomous” action — “going out and executing a series of tasks on your behalf,” Brasca says.

Adopters of AI should think of agents “as an extension of your workforce — as another employee,” he says. That means taking care to define roles, metrics and reporting structures, “in order for them to operate just like humans.”

It’s essential that agentic AI systems be able to explain the reasoning beyond their decisions. Only then can the system reach a level of trust that allows it to function autonomously. “The idea of process governance is super-important,” Brasca says.

Agentic AI “opens up the art of the possible,” he says, but supply chain leaders need to look beyond individual tasks to the larger outcomes they seek to achieve.