Agentic AI: the revolution of an AI that “stops being reactive” and starts deciding

A system of this type, for example, could detect a suspicious email, check if it has reached other employees, automatically block it and alert both those responsible for security and those potentially affected, all without the need for human intervention. These are capabilities that numerous technology providers are already exploring and beginning to incorporate into their solutions, as wholesaler Ingram Micro knows firsthand.

Beyond security, agentic AI also aims to transform areas such as customer service, where it will reduce response times and automate simple incidents; human resources, filtering profiles and organizing interviews; or sectors such as healthcare and finance, relieving human professionals of administrative tasks.

LOOK: NASA will announce the four astronauts of the Artemis III mission on June 9

The goal, according to Hilari, should not be to completely replace people. “The objective should not be to automate everything so that humans do nothing, but rather to take advantage of the capabilities of these agents to multiply our own capabilities as professionals,” he maintains.

A new role for the worker

The expansion of agentic AI will also force us to redefine the role of employees within organizations. In this new model, humans will play a supervisory, validation and coordination role against systems capable of executing complex tasks autonomously.

“The debate should not be whether AI will replace us, but how we reimagine our role in an environment where humans and agents work together,” explains Hilari. In his opinion, “whoever understands that will stop seeing technology as a threat and will start using it for what it really is: a capability multiplier.”

This change will require companies to review their operational structure from the ground up and define the extent to which they wish to grant autonomy to these systems, especially in the early implementation phases. The coexistence between humans and AI agents with the capacity to make decisions that directly affect the business poses new challenges related to governance, supervision and responsibility.

“What is clear is that the window to prepare is narrowing. The companies that begin today to define their governance model, to identify which processes can – and should – be automated and to train their teams to work alongside AI agents, will be the ones that lead the transition. Those that wait for the technology to be a commonly accepted reality will be late,” concludes the Ingram Micro manager.