Agentic AI with Human Oversight Can Play a Major Role in Smart Airport Security, Says Jaideep Mirchandani Group Chairman Sky One

In March this year, the Delhi International Airport introduced the Unified Total Airside Management (UTAM) system, an Artificial Intelligence (AI) – enabled platform designed to enhance the efficiency and safety of aviation operations. It helps in uniting airlines, ground handlers, and airport authorities on a collaborative platform, allowing them to monitor air fleet movements, ground service equipment, and vehicle activities in a more precise manner.

Jaideep Mirchandani, Group Chairman, Sky One

Even as AI continues to gain widespread acceptance in aviation systems worldwide, the industry is also rapidly progressing towards agentic AI. While traditional AI tools mainly offer recommendations based on historical data and signals, agentic AI takes a more proactive role by not only making decisions but also executing them and monitoring outcomes in real time.

Airports like Singapores Changi Airport are exemplifying this shift, and exploring agentic AI to boost operational efficiency, improve customer experience, and reduce the need for human intervention. In India, the first AI-powered airport digital twin was recently launched at Rajiv Gandhi International Airport, Hyderabad. It includes features like intelligent crowd management, flow and queue analytics, real-time operational insights, and optimised backend operations such as smart traffic monitoring.

“The adoption of agentic AI has helped in achieving new standards of customer experience, operational efficiency and making airlines more competitive. Those who adopt them faster will have a competitive edge. Most of these agents are designed to adapt to dynamic scenarios, can respond to passenger feedback and market fluctuations, which helps in revenue optimisation for the carrier while ensuring enhanced experience for the passenger,” says Jaideep Mirchandani, Group Chairman of Sky One.

With the ability to monitor live feeds from terminals, baggage scanners, and access points, agentic tools can detect unusual patterns and anomalies before incidents escalate. “These tools are capable of reading human behaviours by tracking signs such as excessive nervousness and erratic motion, providing an added intelligence layer and flagging potential threats that traditional security systems may overlook. AI agents predict passenger flows, queue congestion, and screening volume, allocating security personnel, adjusting checkpoint openings, and reallocating staff in real time for smoother operations,” adds Mr. Mirchandani.

The AI vision agents can monitor CCTV, detect threats, and alert security teams. They will also support predictive maintenance of aircraft, contributing to safety. However, even as AI agents gain wide acceptance, human supervision remains essential.

“Even if every segment is fully automated, we cannot ignore the necessity of human touch. An AI agent cannot replace human perceptiveness, especially in areas requiring critical decision-making. Both AI agents and human resources, however, will co-exist, complementing each other. While AI tools can assist in various ways, the final resolution may come from humans, offering decisive input and support,” concludes Mr. Mirchandani.