The convergence of 5G and AI at the edge is not just a passing trend; it is a seismic shift that is reshaping our digital landscape. With edge computing emerging as an expanding market projected to soar to a staggering $274 billion in global revenue by 2025, and 5G technology poised to unleash a whopping $3.6 trillion in economic output and 22.3 million jobs by 2035, the potential is nothing short of transformative. This intersection of 5G and artificial intelligence at the edge is not just a game-changer; it is a promise to redefine how we connect, compute, and create value in an increasingly interconnected world.

When we talk about the Edge, we are not just referring to smartphones and IoT devices. We are looking at a range that spans from a couple of servers to dozens of servers in a customer's premises, including warehouse public data centers. Edge computing brings data sources closer, enabling AI processing at the edge to achieve lower latency, higher performance, and lower connection costs. This convergence empowers the 5G network to become more intelligent and meet user requirements, ultimately enhancing the customer experience by providing seamless connectivity from the front end to the back end.

Real-World Applications

The potential of this convergence is immense, with the ability to run mission-critical services across various industries, including retail, public sector, media and entertainment, telco, manufacturing, financial, agricultural, education, and healthcare. For example, in a steel manufacturing factory with high-temperature operations, replacing manual operations with 5G and AI-empowered robots significantly enhances operational efficiency and safety. The following examples highlight the vast potential of AI integration at the edge, driving innovation and meaningful change across industries:

Industry

 Real-World Applications
Retail

Frictionless checkout, inventory tracking, and store analytics

Media & Entertainment

Video tagging, encoding, and regulation compliance

Telco

Modernization of networks for 5G Core, RAN, and local data processing

Financial
Data sovereignty, low-latency trading, and payment processing
Healthcare
AI for imaging diagnostics and patient monitoring
Manufacturing
Quality control, IP and regulations, and autonomous mobile robotics
Agriculture
AI-equipped drones for real-time insights in crop management
Education AI integrated into provision headsets for on-the-spot inferencing
Transportation
Data collection, autonomous driving systems safety

Embracing the Convergence

5G, with its ultra-fast speed, low latency, and massive device connectivity, enables data collection from almost anything. According to industry estimates, the data production is projected to reach yottabytes annually by 2030. This exponential growth in data will be accompanied by a tenfold increase in general computing power and a remarkable 500-fold increase in AI computing power. This brings in the issue of data gravity, where it becomes financially and physically impossible to transmit data to another location after a certain period. Challenges such as scalability, operability, total cost of ownership, sovereignty, privacy, security, and the choice of apps and services create further hinderance.

As organizations adopt these new technologies, they face challenges such as managing hardware and software infrastructure at scale, optimizing costs across multiple vendors, ensuring data sovereignty and privacy, and tapping into an open ecosystem of applications. Private wireless networks, including 5G, can ensure data from front networks can reach AI applications running at the edge, addressing constraints in connectivity that may arise in various locations. The benefits of this convergence are undeniable, offering the potential to transform industries and unlock new possibilities.

Key Considerations

With Gen AI becoming pervasive in our technological landscape, overlooking the edge and radio during management platform and enterprise application design could lead to significant consequences. A thoughtful approach to application design can ensure that the infrastructure can accommodate the convergence. Here are a few other factors to consider:

  • Assess the physical limitations of data center operations and determine the types of hardware needed to support these advancements.
  • Evaluate the GDPR compliance issues, and the need to manage private data within a public compute environment.
  • Utilize AI to collect onsite data and predict flow patterns to proactively allocate resources and enhance performance.
  • Consider the cost implications associated with increasing power consumption by understanding power usage, addressing shortages, and managing peak demands.

Balancing Innovation and Impact

As the demand for AI continues to grow, it is projected that by 2030, AI could account for 3% to 4% of global power consumption. Notably, Google has already reported that AI currently represents 10% to 15% of their power usage, amounting to 2.3 TWh annually. On the brighter side, research from GSMA Intelligence underscores the potential for energy savings through edge processing.

Innovative solutions are being developed to tackle sustainability, data security, and privacy challenges, such as consolidating hyperscaled gateways and attaching storage directly to gateways that connect to local zones in the region. This approach separates the storage requirements and leverages the advantages of public cloud from a compute perspective on a private network, ensuring a highly distributed architecture.

By leveraging AI capabilities, organizations can streamline operations without the need for extensive manual intervention, reducing the reliance on traditional SQL-based approaches. Notably, the ability to process data at the edge without the need to transmit it to the cloud represents a significant energy-saving opportunity. This approach unlocks on-premise capabilities at the edge, contributing to sustainability efforts by optimizing energy consumption and operational efficiency.

Unleashing Potential

Monetization in this space lies in building tailored solutions for edge users, leveraging private networks to run edge applications. It is essential to focus on articulating the tangible value proposition and positive impact on operations, such as revenue generation, cost savings, and operational efficiency. By demonstrating the practical applications and benefits of AI and 5G at the edge, we can unlock unprecedented value for businesses, communities, and individuals.

In conclusion, the convergence of 5G and AI at the edge is reshaping our digital landscape. By embracing this convergence, organizations can tap into its vast potential, transform industries, and unlock new possibilities. It is crucial to consider key factors, balance innovation and impact, and focus on delivering tangible value to unleash the full potential of this transformative synergy.

About the Author

Suruchi Gupta
General Manager, Engineering, Wipro Engineering Edge/Connectivity Practice

Suruchi Gupta is the General Manager and Global Head Technology Development for Wipro Engineering Edge/Connectivity Practice. She is responsible for supporting clients in defining their digital transformation journey, identifying opportunities for business, lead conversations and bring all the capability Wipro and Wipro digital have to offer – which includes positioning/developing and generating revenues for new products and offers with CSPs, NEPs, Hyperscalers, and Enterprises globally.