Beyond the Smartphone
Samsung’s Quiet Bid to Build Europe’s AI Infrastructure
MWC 2026 and the emergence of the AI Infrastructure Stack
Each year, the global telecom industry gathers in Barcelona for Mobile World Congress. For decades the event served primarily as a stage for the evolution of mobile networks and the devices that connect to them. But walking through the halls of MWC 2026, the atmosphere felt fundamentally different.
Smartphones were still present, of course. Yet the deeper narrative unfolding across keynotes, closed-door briefings and operator demonstrations was not about the next device cycle. It was about infrastructure — the invisible architecture that will underpin artificial intelligence, connectivity and digital sovereignty for the coming decade.
In that shifting landscape, one company stood out for attempting something unusually ambitious: building a vertically integrated AI infrastructure stack spanning silicon, networks, devices and industrial systems. The company was not an American hyperscaler or a cloud provider.
It was Samsung Electronics.
From Devices to Infrastructure
For years Samsung has been perceived primarily as a consumer electronics giant. Its Galaxy smartphones dominate global market share charts and its displays and consumer devices are ubiquitous.
But at MWC 2026, Samsung’s messaging pointed in a different direction. The company increasingly presents itself as an infrastructure actor — one operating across multiple technological layers simultaneously.
This repositioning reflects a broader shift in the AI economy itself. Artificial intelligence is no longer confined to cloud data centers. Instead, AI systems are becoming distributed across silicon, networks, edge devices and industrial environments.
In other words, AI is becoming infrastructure.
Understanding this transformation requires moving beyond the familiar narratives of AI software and consumer applications. What is emerging instead is a layered architecture — a stack of interconnected technological systems that together enable AI at global scale.
At Altair, we describe this emerging architecture as the AI Infrastructure Stack.
The Silicon Backbone
At the base of the stack lies the silicon layer — the compute and memory systems that power artificial intelligence.
Here, the conversation increasingly revolves around bandwidth rather than raw computation. As AI models scale, the movement of data between processors and memory becomes the central bottleneck.
“The era of AI demand is shifting from simple computation to massive data movement. Our HBM4 development and advanced 2nm logic process are not just product updates; they are the physical foundation of the AI-RAN ecosystem we are showcasing here in Barcelona.”
Dr. Kye Hyun Kyung
President & CEO, Device Solutions Division
Samsung Electronics
Source: MWC 2026 Technical Keynote – “The Silicon Backbone of Global AI”, March 3, 2026
High-bandwidth memory, advanced process nodes and accelerator architectures form the physical substrate of the AI economy. While companies like NVIDIA dominate the GPU landscape, Samsung occupies a unique position as both a memory leader and semiconductor manufacturer.
But compute alone is no longer sufficient. The next constraint is connectivity.
The Photonics Pivot
Between silicon processors and large-scale AI systems lies another critical layer: interconnect.
As AI clusters scale into thousands of accelerators, moving data between chips becomes as challenging as performing computation itself. Optical networking and silicon photonics are emerging as essential technologies to overcome this limitation.
Companies such as SMART Photonics and Coherent Corp. are developing optical interconnect technologies capable of transmitting vast volumes of data with far greater efficiency than traditional electrical links.
This shift represents what might be called the Photonics Pivot of the AI economy: a transition from electrical interconnects toward optical data movement.
For Europe — home to strong photonics research clusters — this layer could prove strategically significant.
When AI Becomes the Network
Higher up the stack lies the telecommunications layer, which is undergoing its own transformation.
Historically, telecom networks functioned as transport infrastructure: the pipes through which digital information flows. Increasingly, however, those networks are becoming computational environments in their own right.
“We are moving beyond the era where AI is an add-on to the network. With AI-RAN, the network becomes a living, breathing AI entity that optimizes spectrum in micro-seconds. This is the blueprint for Europe’s 6G future.”
Woojune Kim
President & Head of Networks Business
Samsung Electronics
Source: Joint Press Briefing with NVIDIA and Vodafone, MWC 2026
AI-RAN — the integration of artificial intelligence directly into radio access networks — represents a profound shift. Instead of static configurations, networks can dynamically allocate spectrum, manage congestion and predict traffic patterns in real time.
For Europe’s telecom operators, the implications are substantial. As geopolitical pressures reshape supply chains and limit reliance on certain vendors, alternative infrastructure partners become strategically important.
Samsung appears increasingly determined to fill that space.
AI at the Edge
While networks evolve, AI itself is moving closer to the user.
Edge computing and on-device AI are rapidly becoming central to the architecture of digital services. Processing data locally reduces latency, improves privacy protections and decreases dependence on centralized cloud infrastructure.
“True sovereignty in AI requires trust at the hardware level. By processing sensitive data on-device through our Galaxy AI architecture, we provide a localized solution that aligns with European digital values and the AI Act.”
Stephanie Choi
Executive Vice President & Head of Marketing
Mobile eXperience Business
Samsung Electronics
Source: Samsung European Strategy Session, MWC 2026
For policymakers in Brussels, the notion of Sovereign AI — systems aligned with European regulatory frameworks and data protection principles — has become increasingly important.
Edge AI may ultimately prove to be one of the key mechanisms through which those ambitions are realized.
The Industrial Layer
Artificial intelligence will not remain confined to consumer devices or digital services. Its most transformative impact may occur within industrial systems.
Factories, logistics networks and energy infrastructure are gradually evolving toward autonomous operation, powered by AI models operating across sensors, robotics and real-time data flows.
“Our ‘AI-Driven Factory’ initiative is the ultimate stress test for our infrastructure stack. It combines our 5G-Advanced private networks, our custom AI chips and our robotic agents into a single autonomous organism.”
YoungSoo Lee
Executive Vice President & Head of Global Technology Research
Samsung Electronics
Source: Samsung Strategy Paper – “Autonomous Industry 2030”, MWC 2026
In this sense, industrial AI represents the convergence point of the entire infrastructure stack: silicon, connectivity, edge intelligence and robotics operating as a unified system.
The Altair AI Infrastructure Model
The interactions between these layers can be visualized as a strategic architecture rather than a collection of isolated technologies.
Below is the conceptual structure of the Altair AI Infrastructure Model.
This layered architecture reveals why certain companies hold disproportionate strategic influence. Those that operate across multiple layers of the stack gain structural advantages.
Figure — The Altair AI Infrastructure Model: Mapping the layers of the emerging AI infrastructure stack, from silicon and photonics to industrial systems and digital sovereignty.
This layered architecture reveals why certain companies hold disproportionate strategic influence. Those that operate across multiple layers of the stack gain structural advantages.
The Vertical Squeeze
Across the AI ecosystem, companies are increasingly attempting to expand beyond their traditional domains.
Chipmakers push upward into software. Device companies design their own processors. Cloud providers build custom silicon.
Yet few companies operate across as many layers simultaneously as Samsung.
The company manufactures advanced semiconductors and memory. It produces consumer devices and displays. It supplies telecommunications infrastructure and invests heavily in industrial automation.
This creates what might be called a vertical squeeze — a strategic position spanning both the bottom and top of the infrastructure stack.
In an era where technological power increasingly derives from infrastructure control, that positioning may prove decisive.
Europe’s Strategic Moment
For Europe, the emergence of the AI infrastructure stack presents both risks and opportunities.
The continent remains heavily dependent on foreign compute providers and hyperscale cloud platforms. Yet it also possesses strengths in critical areas such as semiconductor equipment, photonics research and advanced manufacturing.
Companies like ASML already occupy indispensable positions in the global semiconductor supply chain. Meanwhile, photonics ecosystems in Eindhoven and elsewhere continue to expand.
If Europe can align these capabilities within a coherent infrastructure strategy, it could play a far more significant role in shaping the global AI economy.
Beyond Barcelona
MWC 2026 offered a glimpse into how the next phase of technological competition may unfold.
Artificial intelligence is no longer merely a software revolution. It is becoming an infrastructure race — one involving silicon, optical networking, telecommunications, edge devices and industrial systems.
Seen through that lens, Samsung’s presence in Barcelona looked less like a smartphone showcase and more like a quiet declaration of intent.
Not simply to build the devices of the AI age.
But to help construct the infrastructure beneath it.
Image: AI-generated illustration
Inside Samsung’s pavilion at MWC in Barcelona, where the company presented its Galaxy AI ecosystem alongside next-generation telecom infrastructure.
This article is part of Altair Media’s special coverage of Mobile World Congress 2026.
Follow ongoing analysis and reporting on the strategic shifts shaping global connectivity on our dedicated page: The Future of Connectivity — MWC 2026
