ChatGPT generated image by The Tech Panda
Telecom operators are entering a pivotal phase, where their future role in the AI economy is no longer guaranteed but must be actively built. Armed with control over national infrastructure, regulated data environments, and expanding edge compute assets, telcos are positioning themselves as sovereign AI powerhouses rather than mere connectivity providers. From billion-dollar AI factories and GPU-optimized data centers to satellite partnerships and on-device intelligence, the industry is undergoing a structural shift.
However, can operators translate these advantages into full-stack AI leadership, or risk yielding value to hyperscalers dominating the higher layers of the stack?
From billion-dollar AI factories and GPU-optimized data centers to satellite partnerships and on-device intelligence, the industry is undergoing a structural shift. However, can operators translate these advantages into full-stack AI leadership, or risk yielding value to hyperscalers dominating the higher layers of the stack?
As per reports, telcos in the future can cash in well from the rise of sovereign AI. Their advantage? They operate national infrastructure, manage regulated data environments, and increasingly control edge compute assets.
However, can they convert those structural advantages into credible AI leadership, rather than simply transporting someone else’s models?
As per Juniper Research data, we can expect news around domestic data centers optimized for GPU workloads, alongside efforts to assemble vertically integrated AI stacks across compute, orchestration, hosting, and security. For example, in Singapore, Singtel RE:AI launched as the nation’s first sovereign AI factory, offering high-performance NVIDIA GPU clusters. Similarly, Orange (France) uses its “Orange Business” arm to provide end-to-end AI consulting, hosting on their sovereign Bleu cloud (built with Microsoft), and managed security services to ensure AI outputs meet EU AI Act requirements.
These operators are positioning themselves not just as connectivity providers, but as full-stack AI enablers. That includes running inference at the edge, hosting enterprise AI workloads, and offering governance-aligned environments for regulated sectors. However, lacking sovereign AI capabilities, operators risk being relegated to utility providers while hyperscalers capture the high-value layers of the AI economy.
In 2026, Tier 1 operators transitioned from experimental pilots to multi-billion dollar infrastructure commitments. These investments focus on high-density GPU-optimized data centers and the creation of “AI Factories” to provide sovereign cloud services for regulated sectors.
In Germany, Deutsche Telekom announced a €1 billion ($1.15 billion) partnership with NVIDIA to launch an “Industrial AI Cloud” by Q1 2026. The investment includes a high-security, liquid-cooled data center in Munich equipped with 10,000 NVIDIA Blackwell GPUs. There is also a tie up with SAP on the “Deutschland Stack” to provide sovereign digital infrastructure for public institutions and manufacturers.
In South Korea, SK Telecom unveiled an “AI Native” strategy at the MWC 2026, involving a multitrillion-won overhaul of its core infrastructure. The plan involves 1GW-class hyperscale AI Data Centers (AIDC) across Korea to establish the nation as Asia’s primary AIDC hub and investing in its proprietary sovereign AI foundation model, scaling it from 519B to over 1 trillion parameters to support domestic industry-specific solutions.
Satellite network operators are moving from fringe presence to central players. Starlink/SpaceX is integrating its Low Earth Orbit (LEO) constellation with terrestrial networks through a strategic, collaborative approach, transitioning from a solely direct-to-consumer internet provider to a key “network-of-networks” infrastructure provider.
At the Mobile World Congress (MWC) 2026, SES positioned its multi-orbit strategy as a resilient, high-performance alternative for connectivity, featuring direct-to-device (D2D) capabilities through partnerships with companies like Lynk Global. Azercosmos is also focusing on expanding its own satellite communication and IoT services.
Operators are zeroing down on the commercial side of non-terrestrial networks (NTNs). So, will these NTNs become money making parts of their portfolio?
D2D is starting to dominate the conversation. The integration of satellite connectivity into consumer smartphones and IoT devices is evolving rapidly, positioning NTNs primarily as a complement to terrestrial coverage, bridging the “last 4%” of geographical gaps rather than replacing cellular networks.
The Apple iPhone (14–17 series) uses the Globalstar network to enable Emergency SOS and, with iOS 18, satellite messaging, while the Google Pixel (9 and 10 series) integrates Skylo for Satellite SOS and texting, and Samsung’s Galaxy S25 and S26 series support satellite-based communication through partnerships like T-Mobile’s Starlink-powered “T-Satellite,” allowing users to send messages even without cellular coverage.
The position of AI as core to network development is increasing, from performance optimization to cost efficiency. What remains to be seen is where the compute will sit. Considering latency is an issue with many use cases, AI has to sit closer to the edge, maybe even on the device itself. This shift is a big one and operators as well as device manufacturers will make it happen.
The Mobile World Congress (MWC) 2026, showcased how on-device AI transitioned from a niche feature like smart cameras to a foundational element of mobile architecture. Manufacturers and chipmakers focused on local processing to deliver low-latency, privacy-first experiences that operate without cloud reliance.
Read more: Striking a balance between digital independence & online safety for kids
Mobile platforms like Snapdragon 8 Elite Gen 5 (Qualcomm) powers flagship devices like the Samsung Galaxy S26 Ultra. It features the Hexagon NPU, which enables agentic AI interfaces that can anticipate user needs across messaging and navigation.
Devices like Samsung Galaxy S26 Series showcased “anticipates intent,” using Galaxy AI to link activity across apps (e.g., automatically adding an email travel confirmation to a calendar).
Physical AI is another tech to watch out for, which has emerged as a major trend in the telecom industry in 2026. The MWC buzzed with ‘embodies intelligence’, from Ericsson and Nokia demonstrating how private 5G and intelligent edge architecture allow robots to offload complex “brain” tasks to the network, making them lighter, cheaper, and more energy-efficient, to T-Mobile and NVIDIA tying up to turn base stations into distributed AI computers that can understand the logic of the physical world in real time.
In 2026, the smartphone market has shifted from an “app-centric” model toward an “agent-centric” architecture, where the operating system itself acts as an autonomous digital worker. These new models move away from isolated apps in favor of an Agentic UI that prioritizes user intent and cross-service execution.
Those that successfully integrate compute, connectivity, and intelligent services into cohesive platforms will emerge as key orchestrators of the AI-driven economy. Those that don’t, risk being reduced to utility providers in a market increasingly defined by who owns the intelligence, not just the network.
For example, Nubia M153 (“Doubao Phone”) developed by ZTE in partnership with ByteDance, is considered the first fully agentic smartphone. The Samsung Galaxy S26 Series features Samsung’s third-generation AI. Then there is the Honor Robot Phone, which features a physical three-axis gimbal camera and is built around Augmented Human Intelligence (AHI).
The telecom industry is at an inflection point where infrastructure alone is no longer enough. Sovereign AI, non-terrestrial networks, and agentic, on-device intelligence are redefining what it means to be a network operator in 2026. Those that successfully integrate compute, connectivity, and intelligent services into cohesive platforms will emerge as key orchestrators of the AI-driven economy. Those that don’t, risk being reduced to utility providers in a market increasingly defined by who owns the intelligence, not just the network.
In global financial system, while traditional banks once dominated the flow of credit and capital,…
Against the backdrop of a resilient macroeconomic environment and sector-specific growth dynamics, salary increment budgets…
The Tech Panda takes a look at recent funding events in the tech ecosystem, seeking…
For years, many founders believed that successful fundraising was about building the perfect pitch deck.…
The Tech Panda takes a look at recent launches in the superfast field of Artificial…
As the growth prospects of economies around the world battle against skyrocketing costs, geopolitical instability,…