Artificial IntelligenceInterviews

Beyond Traditional AI: Agentic AI Emerges

Manoj Ganapathy, Division Manager for Visitors Management & Robotics Division at Jacky’s Business Solutions explains how Agentic AI stands apart from older systems, its ability to tackle complex, multi-step tasks, and the importance of balancing AI independence with human control

How do you define Agentic AI, and how does it fundamentally differ from traditional chatbots or rule-based automation?
Agentic AI represents a foundational evolution in how artificial intelligence systems operate. Unlike chatbots or rule-based tools that execute pre-programmed commands, agentic AI systems are autonomous, goal-driven, and capable of planning and adapting in real time. These systems understand high-level objectives, decompose them into actionable steps, and dynamically adjust strategies based on evolving conditions.

At Jacky’s Business Solutions, we see Agentic AI as the core of next-generation enterprise automation—systems that act with purpose, learn from outcomes, and interact fluidly with both digital and physical environments. Where traditional automation reacts, agentic systems reason, self-correct, and proactively drive outcomes that align with broader organisational goals.

Beyond task automation, what complex, multi-step problems are agentic AIs uniquely positioned to solve that current AI models cannot?
Agentic AI excels in scenarios where complexity, change, and coordination converge. These systems can orchestrate multi-stage processes—such as adjusting logistics in real time, personalising large-scale customer interactions, or dynamically reconfiguring operations in response to external disruptions.

Unlike conventional models that focus on pattern recognition or isolated tasks, agentic systems engage in long-term planning and iterative learning. For example, they can autonomously manage the flow of people and information in large facilities, continuously optimising for efficiency and compliance. In our deployments across the GCC, we’re seeing these capabilities emerge in areas like facility automation, smart retail, and autonomous visitor engagement.

How much autonomy should Agentic AI have in decision-making, and where should humans remain in the loop?
The level of autonomy granted to Agentic AI should depend on the risk, complexity, and ethical implications of the task. For routine, operational decisions—especially those with clear parameters—AI agents can act independently to drive efficiency and speed. However, in high-stakes contexts involving safety, privacy, or regulatory implications, human oversight remains essential.

At Jacky’s, we advocate for a layered autonomy model. Agentic AI should take the lead where it excels—processing data, executing decisions at scale—but always within guardrails designed to ensure transparency, accountability, and alignment with human values. Humans must retain authority over strategic, ethical, and context-sensitive decisions.

What infrastructure challenges (such as power, cooling, latency) arise when deploying Agentic AI at scale?
Agentic AI systems place substantial demands on infrastructure. They require sustained access to compute power, large memory bandwidth, and low-latency connectivity—often across edge, cloud, and on-premises environments.

In the GCC, environmental factors add complexity. High-density compute clusters generate significant heat, necessitating advanced liquid cooling and efficient power management. Latency is another key constraint—especially for real-time applications like autonomous agents in healthcare or logistics. This is driving a move toward edge computing, where inferencing occurs closer to the data source, reducing delay and improving responsiveness.

To address these challenges, Jacky’s and our partners are investing in modular, hybrid infrastructure models—balancing cloud scalability with local performance to support mission-critical deployments.

What’s the next frontier for Agentic AI—will we see AI “agents” collaborating like human teams?
Absolutely. The emerging phase of Agentic AI involves multi-agent collaboration—where systems don’t just operate autonomously, but also interact, negotiate, and coordinate like distributed teams. These agents can dynamically divide tasks, manage dependencies, and adapt their roles as goals evolve.

We believe this is where the most exciting business transformation will occur. Whether it’s retail bots managing end-to-end inventory or digital agents streamlining government service delivery, collaborative AI will amplify efficiency and resilience. However, it also raises new questions about governance, interoperability, and shared objectives—areas where robust design and oversight are crucial.

Will open-source models keep pace with proprietary Agentic AI systems, or will there be a widening gap?
The ecosystem is evolving rapidly. Open-source models—like Meta’s LLaMA and Google’s Gemma—are closing the performance gap, offering enterprises cost-effective, transparent alternatives that can be fine-tuned for local or sector-specific needs. They also foster broader innovation, especially in markets like the Middle East that value adaptability and sovereignty.

However, proprietary systems still lead in turnkey capabilities, fuelled by exclusive datasets, massive compute resources, and vertical integration. We anticipate a hybrid future—where enterprises use open-source platforms for experimentation and customisation, while leveraging proprietary tools for enterprise-grade performance and integration. At Jacky’s, we help clients balance both, ensuring they get the right mix of innovation, control, and scalability.

Show More

Chris Fernando

Chris N. Fernando is an experienced media professional with over two decades of journalistic experience. He is the Editor of Arabian Reseller magazine, the authoritative guide to the regional IT industry. Follow him on Twitter (@chris508) and Instagram (@chris2508).

Related Articles

Back to top button