Who is watching the AI agents?
For better or worse, AI agents are now a part of the workforce. They write code, analyze documents, respond to customers, coordinate workflows, and make decisions across multiple business systems with very little human involvement. AI agents have proven they can do the work. Now, enterprises must prepare for a world where they do far more of it. With Gartner predicting that the average Fortune…
AI agents, once a fringe technology, have now become integral to the modern workforce. They execute tasks ranging from coding and document analysis to customer service, workflow coordination, and decision-making with minimal human oversight. As per Gartner's forecast, by 2028, the average Fortune 500 company will operate over 150,000 AI agents, necessitating robust governance measures.
Traditional governance models, built around human systems, no longer suffice due to the rapid pace of AI deployment and the sheer volume of activities generated by multiple agents. A survey by IBM revealed that 70% of technology executives believe AI deployment outpaces their ability to track it, while Gartner found that only 13% of organizations consider their AI governance sufficient.
Deloitte's findings echo this, with 21% of enterprises reporting mature governance for agentic AI. The solution lies in redesigning governance to integrate AI at its core. Guardian agents form a new category of AI tasked with supervising the work of business agents. These agents monitor interactions with enterprise systems, evaluate planned actions against organizational policies, and intervene when policies are breached.
They can generate audit trails, alert human reviewers, or block actions in high-risk situations. Unlike traditional guardrails that address anticipated risks, guardian agents enable organizations to manage risks proactively, aligning governance speed with AI deployment. While safeguards and approvals are crucial, they can only prevent risks anticipated by policy creators.
Human judgment, which considers business context and ethical considerations, remains essential. Guardian agents facilitate this judgment by applying it consistently across every AI interaction, enhancing governance scalability without replacing human accountability. As AI agents evolve at an unprecedented pace, the decisions organizations make about their operation will significantly impact whether these technologies yield value or pose risks. Blake Brannon, chief innovation officer at OneTrust, underscores this evolving landscape.
Written by urgent.news from Fast Company's reporting — not their text. Machine-written — may contain errors; check the original before relying on it.
