AI + 5G: From 'Multiplicative Buzzwords' to 'Foundations for Results'
AI + 5G: From 'Multiplicative Buzzwords' to 'Foundations for Results'
Real transformation in the telecommunications industry, pioneered by the edge
Entering 2026, the terms "AI" and "5G" are once again garnering attention in the telecommunications industry. However, their relationship is no longer just a combination of technical trends. How to connect AI and 5G to "real-world results" is becoming a common theme for the entire industry. A discussion that symbolizes this theme took place recently, moderated by Abe Nejad of The Network Media Group (NMG). In this session, the perspectives of telecommunications operators and solution providers were used to delve into how AI + 5G + Edge can generate concrete value. What was particularly impressive was the redefinition of AI not as a "standalone use case," but as a "foundational layer" for producing results.
Speakers:
Bhupinder Singh (President, APAC, Vodafone)
Faisal Ghazaleh (Global Head of Solutions – OSS, Rakuten Symphony)
Video Description: The fusion of AI and 5G is shaping real-time, data-driven enterprises. Leaders from Rakuten Symphony and Vodafone delve into how edge computing, connectivity, and "application intelligence" are driving outcome-focused innovation and redefining industries across the APAC region.
Thinking from an "outcome-first" perspective rather than a technology-first one
At the heart of the discussion was the concept of "Outcome-back." This is an approach that moves away from technology-first thinking—such as "introducing it because AI is available" or "utilizing it because 5G exists"—and instead defines KPIs that are meaningful to companies and society, then designs the technical configuration by working backward from there. For example, corporate KPIs include the following:
・Reduction in latency
・Reduction in failures and accidents
・Improvement in throughput
・Optimization of energy consumption
To achieve these results, AI alone is insufficient. The discussion concluded that three elements must be established simultaneously to create value. Only by being cross-functionally integrated into all three layers can AI become something that "delivers results."
Element 1: Connectivity: High-speed, low-latency communication via 5G
Element 2: Processing Power: Immediate computation via GPUs and edge computing
Element 3: Application Layer: Software responsible for business logic and decision-making
The missing piece: Edge computing
In this discussion, edge computing was emphasized as playing a particularly important role. While 5G enables high-speed communication, sending data back and forth to the cloud results in a loss of real-time capability. In use cases requiring millisecond-level decisions, this latency can be fatal. This is where an architecture that processes data at the "edge" of the network comes in. The following cases were introduced as specific examples. These are all use cases that can only be realized when 5G bandwidth, edge processing, and AI judgment are combined. The edge can be called the "bridge" that converts the "speed" of 5G into the "actions" of AI.
・Instantly analyzing vehicle sensor data at the edge to issue warnings the moment danger is detected
・AI judging wireless network usage in real-time to dynamically reallocate radio resources
・Achieving double-digit percentage energy savings through intent-based algorithms
First deep vertically, then expand horizontally
Realistic views were also presented regarding the areas where AI + 5G adoption will proceed first.
Initial adoption will progress in environments such as the following:
・Closed and managed networks
・Industries with strict regulatory and compliance requirements
・Areas where ROI can be clearly measured
The flow is for "vertical-specific" adoption to lead, followed by horizontal expansion into the mid-market and broader fields. A challenge mentioned here is economic viability for the mid-market. Even if technically possible, adoption will not progress if the cost structure does not align. It was stated that how this is designed will be the key to the next growth phase.
Trust as a "prerequisite," not an "added value"
As AI and telecommunications become deeply embedded in social infrastructure, the importance of security, privacy, and ethics is rapidly increasing. In many countries and regions, data is treated as a "national resource." Therefore, the following points were emphasized:
・Architecture design that respects data sovereignty
・Transparency and traceability of AI models
・Alignment with standardization bodies and regulations
・Governance for new models, including agentic AI
The message presented here is clear: "Trust" is no longer a differentiator; it is the operating system itself. Without trust, no matter how advanced the AI, it will not be accepted by society.
Beyond corporate value, toward social value
Toward the end of the session, it was touched upon that the potential of AI + 5G extends beyond corporate ROI.
・Improvement of public safety
・Disaster prediction and early response
・Environmental monitoring
・Strengthening of infrastructure resilience
These are all extensions of the same technology stack discussed previously. Energy optimization technology for OPEX reduction can also be repurposed for protecting human life and the environment. This point was emphasized, showing that AI + 5G is evolving from a "cost-cutting tool" into a "foundation for social contribution."
Key points that emerged from the session
Summarizing this discussion, it can be condensed into the following five points:
・Outcome-first, not technology-first: Define business and social results first
・The edge is an essential element: Real-time capability cannot be established without local processing
・From vertical expansion to horizontal expansion: Limited areas first, then scale
・An era where trust becomes the OS: Security and ethics are minimum requirements
・From cost reduction to social contribution: The same mechanism enhances sustainability
To turn speed and intelligence into action
Finally, the words of Faisal Ghazaleh of Rakuten Symphony succinctly express the essence of this session: "What we need is not technology for the sake of technology, but a strategy that starts with what we want to deliver to the customer. It is important to integrate AI into the three aspects of connectivity, processing, and applications to realize truly meaningful real-time visualization, monitoring, and 'action'." AI and 5G are no longer just future concepts. By gaining a realistic foundation called the edge, they are beginning to evolve from "usable technology" into "mechanisms that produce results." This session clearly indicated that turning point. The next axis of competition in the telecommunications industry is shifting from "how fast" or "how smart" to how much one can move reality.
Personal opinion and analysis: "AI + 5G + Edge" is not a technological revolution, but a battle over the control of decision-making
The fusion of AI and 5G has been talked about many times as the "next revolution." However, to be honest, it is a fact that for the past few years, expectations have run ahead of actual results in the field. What was suggestive about this discussion was that it presented a perspective to finally fill that "gap period." It was not about increasing new technologies, but about facing head-on the problem of where to place the location and speed of decision-making.
AI is not a smart tool, but the organizational structure itself
Many companies have misunderstood AI as a "convenient analysis tool" or an "extension of automation." However, the image of AI discussed in this session is clearly different from that. AI is no longer a standalone product. It is an entity that rewrites network design, operational philosophy, and business processes themselves from the inside. What is important here is not "whether to introduce AI," but the issue of governance: which KPIs, at what timing, should be entrusted to AI rather than humans. This is not just a matter for the technical department; it is approaching the level of management decision-making itself.
The essence of the edge is not low latency, but sovereignty
Edge computing is often explained as a "mechanism for achieving low latency." However, what is more essential is the location of sovereignty. In a cloud-centric model, there are structural problems: data goes outside, decisions are made far away, and the location of responsibility becomes ambiguous. The edge is different. Data stays on-site, decisions are completed on-site, and accountability can be clarified. This is not just a difference in architecture, but a difference in philosophy regarding "who controls the world in real-time." That is why the edge is rapidly increasing in importance in the context of data sovereignty, national sovereignty, and corporate sovereignty.
Outcome-back is not an ideal, but the only realistic solution
The term "Outcome-back" (outcome-first) may sound like a platitude at first glance. However, looking calmly at the current state of the telecommunications industry, this is not an ideal, but a survival strategy. This is because there is a reality where technology is becoming commoditized, differentiation has limits in speed and price, and AI investment is expanding without limit. Companies that cannot ask the question, "Which AI is needed to produce this result?" rather than "What can be done with this AI?" will turn their AI investments into a "cost swamp."
Vertical expansion leading is not because of being "timid"
Some view the fact that adoption is progressing from vertical markets as a strategy that is too cautious. However, I believe this is extremely rational. The reason is clear: outcome indicators are clear, the regulatory environment is well-defined, and it is easy to explain the return on investment. In particular, telecom x AI x edge is a field where the cost of failure is too high. That is why it is necessary to first create a "model of success" in a controllable environment. The problem is not vertical expansion itself, but whether one has the design philosophy to expand horizontally from there from the beginning.
Trust will become a licensing system, not an added value
In this discussion, the most significant meaning lies in the idea that "Trust is the operating system." In other words, this is a declaration that an AI or network that cannot demonstrate trust does not have the qualifications to participate in society in the first place. From now on, AI whose data origin cannot be explained, whose decision-making process cannot be tracked, and whose ethical boundaries are ambiguous will become something that should not be used, regardless of whether it is usable or not. This is a matter of social consensus rather than a matter of regulation.
The fact that OPEX reduction and social contribution are on the same line
What is very important is that initiatives such as energy optimization and operational automation simultaneously satisfy different value axes: cost reduction, sustainability, and public safety. Herein lies the true strength of AI + 5G + Edge. This is not a story about "companies doing good things." It is a structure where social value is naturally born as an extension of rationality. Only companies that can create this structure will be supported in the long term.
Not telecommunications, but civilization design
Ultimately, this theme is not about telecommunications technology. Where to decide, who controls, and at what speed to move the world. This is the design philosophy of digital civilization itself. AI and 5G are merely means to that end. The edge is the place to anchor that philosophy in reality. Future competition will shift not to who has the fastest network or who has the smartest AI, but to who can make the most correct decisions, in the place closest to reality, with the most responsibility. I think this discussion clearly indicated that turning point.
AI + 5G at the edge: From hype to shaping real outcomes in telecom
https://symphony.rakuten.com/blog/ai-5g-at-the-edge-from-hype-to-shaping-real-outcomes-in-telecom
