Dominic Jainy is a seasoned IT professional who has spent his career at the intersection of artificial intelligence, machine learning, and blockchain infrastructure. As the telecommunications industry navigates a period of stabilization, Dominic provides a vital perspective on how software-defined layers are replacing traditional hardware to meet the demands of a modern, automated world. His insights help clarify the shift from simple connectivity to a future where programmable networks serve as the backbone for intelligent machines and enterprise-level automation.
How is the transition from hardware-centric SIM products to software-defined infrastructure layers fundamentally changing the way we think about global connectivity?
We are currently witnessing a massive logic-shift where connectivity is no longer viewed as a static hardware commodity, but as a fluid, software-defined infrastructure layer. A Danish IoT provider is currently making waves with a €100 million war chest specifically designed to rip up the traditional IoT rule book and push this agenda of programmability. By moving away from traditional SIM products and toward standards like SGP.32 and eUICC, we can now spin up high-end, on-demand services with a level of dynamism that was previously impossible. This isn’t just a niche trend for a few progressives; it is a discipline shift that allows for the seamless integration of physical AI and robot automation into our daily machine networks.
With the RAN market expected to remain relatively flat from 2026 to 2030, how are operators justifying continued investments in AI-RAN and massive MIMO?
While it is true that the post-5G correction has ended and the market is projected to be broadly flat through 2030, operators are focusing on internal efficiencies rather than simple market growth. Companies like Orange, Optus, and SK Telecom are actively testing practical AI use cases to ensure their existing infrastructure is as lean and effective as possible. We are seeing AI-RAN, cloud RAN, and massive MIMO grow within this flat financial envelope because they offer a way to manage networks more cheaply and easily. The strategy right now is to maximize the performance of every existing asset, preparing the ground for 6G without necessarily expanding the total capital expenditure.
Could you elaborate on the distinction between using AI to optimize current networks and building networks specifically to accommodate the traffic patterns of future AI agents?
This is a crucial distinction that defines our current erthe difference between AI for RAN and RAN for AI. AI for RAN is the immediate, practical application where we use machine learning to make networks more efficient, easier to operate, and significantly more cost-effective. On the other hand, RAN for AI represents the industry’s “grand wager,” a long-term bet that new traffic patterns from autonomous machines and AI agents will create massive new demand. This second path is what will eventually justify the need for increased capacity and more spectrum as these digital agents begin to dominate our network usage. We are already seeing the groundwork for this with the rise of 800G and 1.6T optical testing demands to support that eventual surge in data.
Where do you see the most significant influx of new capital entering the telecommunications space given the maturity of public consumer networks?
The most exciting growth is happening in the private wireless sector, where enterprises are bringing genuinely new capital into the market to support their specific industrial needs. We are seeing organizations like OCUDU receive new funding specifically for factory demos, proving that the demand for high-performance machine networks is very real in the manufacturing sector. Additionally, the acceleration of AI networking is being fueled by significant investments, such as Ciena’s recently launched $200 million venture fund. These enterprise-driven projects are moving beyond simple connectivity, focusing instead on how private 5G and physical AI can revolutionize efficiency on the factory floor and beyond.
What is your forecast for the role of AI agents in network demand over the next few years?
My forecast is that we are rapidly approaching the “agentic world” where the difference between human-driven and machine-driven traffic becomes the primary concern for network architects. As leaders at Verizon have pointed out, there is a clear line being drawn where AI agents and automated systems start to dictate the flow of data across our global infrastructure. Over the next few years, I expect these autonomous agents to create a surge in demand that finally validates the massive infrastructure investments we’ve been making. We are moving toward a reality where the network is not just a pipe for information, but a sophisticated environment designed specifically to support the high-frequency, high-reliability needs of an agentic economy.