Living On The Edge: Where Physical And Digital Worlds Intertwine

Forbes

09-07-2025

Javed Khan, President of Software, Advanced Safety and User Experience.
The manifestation of artificial intelligence (AI) is undergoing a fundamental shift. Already, we're moving beyond the theoretical capabilities of AI in the cloud to explore its tangible impact on the physical world. Increasingly, the most transformative AI applications are emerging not just in centralized data centers, but also at the intelligent edge—the point where real-time sensing, decision making and action converge within integrated systems.
Consider the implications. From vehicles capable of navigating complex environments autonomously to industrial facilities that can dynamically adjust operations based on real-time data, edge-based AI is rapidly becoming the new benchmark for industries where speed, precision, safety and immediate responsiveness are critical.
This isn't a gradual technological evolution; it signifies a profound rethinking of where intelligence resides and how it can deliver tangible value on a massive scale.
The Intelligent Edge In Action
Currently, examples of edge-based AI include familiar applications like facial and fingerprint recognition on our mobile devices, or even more complex autonomous systems such as industrial robots or delivery drones.
While edge AI might sound abstract, its practical implementation resembles our own human body's intelligence system. Our body uses sophisticated sensors—our five senses—to capture environmental data: eyes detecting movement and light, skin sensing temperature, ears picking up sound vibrations and more. These sensory inputs connect to our brain—nature's powerful yet energy-efficient processing unit—which performs local computation without needing to consult external intelligence.
Our learned behaviors and reflexes function like lightweight AI models, optimized for instant on-device execution rather than requiring conscious thought. Underpinning these components is our autonomic nervous system—similar to specialized edge operating systems—designed for real-time processing of critical functions. Finally, our ability to learn from others and update our knowledge through communication mirrors the cloud orchestration aspect, enabling seamless connectivity and continuous updates across our distributed neural network.
This model is the functional blueprint that inspires advanced driver-assistance systems (ADAS) on the road today.
In a typical ADAS cycle, AI models function much like our human learning process. Just as we develop core skills through childhood experiences, these AI models are trained on extensive datasets before being deployed to the vehicle's sensory organs—radar and cameras that serve as the car's eyes and ears.
Our brain constantly processes new sensory information to refine our reactions and decision making. ADAS performs similarly by continuously gathering real-world driving data. This information flows back to update the core intelligence, similar to how our experiences shape our neural pathways. The improved understanding is then distributed throughout the fleet—comparable to how humans share knowledge through communication and learning from others' experiences. This creates a closed-loop system of continuous learning and enhancement that mirrors our own lifelong adaptation to new situations and environments.
As humans successfully adapt and transfer cognitive models across diverse situations—applying skills learned in one context to entirely different scenarios—the fundamental principles and technological components that enhance safety in vehicles can also be applied to diverse sectors such as healthcare, telecommunications, industrial manufacturing and robotics.
Truly, the underlying need for the rapid processing and localized intelligence unlocked by edge AI transcends specific industry applications.
Get A Head-Start In Edge AI
For enterprises and organizations aiming to leverage the potential of AI at the intelligent edge, I've found that the following strategic steps can ensure a significant advantage:
• Identify key applications. Focus on scenarios where real-time insights and autonomous actions can yield substantial gains in efficiency, safety or customer/user experience. Consider applications in industrial automation, predictive maintenance or enhanced situational awareness in complex settings.
• Invest in infrastructure. This includes selecting edge hardware with adequate processing capabilities and energy efficiency, along with reliable sensor technologies tailored to the specific use case.
• Develop or upskill AI/ML expertise. Creating and deploying efficient AI models for edge devices requires specialized skills in areas such as model optimization and inference.
• Implement a management platform. This is key for scalability and maintainability. When vetting platforms, look for those that can deploy and upgrade AI models via over-the-air (OTA) updates, monitor device performance and ensure the security of edge deployments.
A well-defined orchestration strategy will facilitate the effective management of a growing number of intelligent edge devices and use cases.
The Real-Time Revolution At The Edge
The intelligent edge represents a significant evolution in how we interact with and utilize AI. It's not only about processing data closer to its origin; it's about enabling what can be described as "physical AI"—intelligent systems capable of perceiving, reasoning and acting within the physical environment in real time. This interconnectedness is foundational to the next wave of innovation across numerous industries.
For example, in the context of modern vehicles, this entails more than radars that simply detect or recognize surroundings. It invokes a deeper, human-level of understanding, such as the ability to differentiate between similar objects in challenging conditions, to anticipate the behavior of pedestrians or adapt vehicle operation based on driver cues. These nuanced, context-aware decisions necessitate rapid processing and cannot rely on constant communication with a remote cloud.
Ultimately, the need for real-time intelligence at the edge is driven by the imperative for immediate and safety-critical actions.
While cloud computing will continue to be essential for AI development and data management, I believe the next significant advancements in AI-driven value will emerge at the edge. Looking ahead, the continued evolution and proliferation of the intelligent edge may democratize AI adoption across a wide spectrum of industries, inspiring innovative applications that we can only begin to imagine today.
Specifically, intelligent edge systems capable of sensing, deciding and acting in real time—in the real, physical world—are poised to unlock substantial improvements in efficiency, insights and safety. The foundational work being done to establish robust and scalable intelligent edge architectures in demanding environments is creating a transferable blueprint for broader transformation.
By embracing the intertwined principles of localized intelligence and real-time AI processing, organizations across diverse sectors can achieve smarter, safer and more insightful operations, paving the way for a truly intelligent, interconnected and more efficient enterprise.
Forbes Technology Council is an invitation-only community for world-class CIOs, CTOs and technology executives. Do I qualify?