Edge Computing in the AI Era

The computing market is currently undergoing a form of correction. A growing realization is emerging from a “cloud-only” push strategy: some processing must return to the ground and move closer to the data source – to the edge. This phenomenon, known as Edge Computing, is not a passing technological trend but a significant economic and technological growth driver, as highlighted in a recent IDC report*.
According to forecasts, global spending on edge computing solutions is expected to reach approximately $380 billion by 2028, reflecting an impressive compound annual growth rate (CAGR) of 13.8%. This growth has deep technical implications for designing, developing, and managing information systems and compute-based products.

The growing need for real-time data processing, as close as possible to the point of data creation or consumption, is one of the main factors driving the edge computing market. Edge computing is particularly prevalent in applications such as autonomous vehicles, intelligent video analytics, industrial control systems, and medical devices.
These use cases demand substantial pre-processing or extremely fast response times, which are not feasible when data must be sent back and forth to a remote data center or the cloud. The increasing proliferation of IoT devices and smart systems further amplifies the data generated at the edge, necessitating localized processing solutions.

Beyond that, the integration of Artificial Intelligence (AI) with edge computing opens up new possibilities. The ability to run AI models directly on edge devices enables fast and decentralized decision-making, reduces dependency on constant cloud connectivity, saves bandwidth, and is more cost-effective than relying solely on cloud processing. A notable trend supporting AI at the edge is the integration of dedicated AI accelerators into edge systems or using processors with built-in AI acceleration, such as NVIDIA® Jetson™.

These processors enable the local execution of small language models and complex multi-modal models like Vision/Language Models (VLMs). The continuous and rapid improvement of multi-modal models – combining different data types such as images, text, and audio – enables powerful on-device capabilities that were not possible just a few years ago. Beyond processing power, privacy and data security requirements and various regulatory constraints often mandate the local processing and storage of sensitive information.

From a technical perspective, spending forecasts point to several key directions. In the near term, the most significant investments will focus on hardware, driven by the need to deploy more powerful edge devices equipped with advanced processing and AI capabilities. However, as the market matures, IDC projects that investment in services will surpass hardware by 2028. This shift reflects the increasing complexity of deploying, managing, and maintaining distributed edge infrastructures at scale. Organizations will increasingly require orchestration services, specialized cybersecurity, and the development of applications tailored for edge environments.

A wide range of industries are adopting edge computing solutions. According to IDC, the retail and high-tech sectors, along with medical products and life sciences, are leading the way in investment. In retail, edge computing enables real-time customer behavior analysis, intelligent inventory management, and enhanced customer experience through autonomous checkout systems and AI-based recommendation engines. In industrial settings, it supports a new generation of vision-based testing systems. In cybersecurity, edge computing combined with AI allows real-time image and video analysis. In the medical device sector, edge products with AI acceleration will enable the interpretation of raw medical data with unprecedented accuracy and speed.

IDC’s report emphasizes edge computing as a vital technological and economic force for the coming decade. The sector’s remarkable growth is fueled by increasing business needs for real-time processing, advancements in AI capabilities at the edge, and the evolution of communication infrastructure. The technical implications include a shift to distributed architectures, the need for advanced tools to manage and orchestrate edge environments, new security challenges, and a demand for applications optimized for resource-constrained environments. A deep understanding of these trends and investment in edge technologies is critical for organizations aiming to remain relevant and competitive in a rapidly evolving digital landscape.

**Yoel Jacobson is the CTO at HIPER Global**

* Reference – An article published on March 17, 2025 by the “International Data Corporation” (IDC), which released its latest forecast for Worldwide Edge Computing Spending Guide, featuring a new enterprise industry taxonomy

Based on the article that was published in the “P&C” magazine on May 21, 2025