Article 1 | LCDC
Written in collaboration with Prof. Thomas Bohné and Dr. Sławomir Tadeja from the Cyber-Human Lab (CHL) at the University of Cambridge.
March 26th, 2026
In the coming years, augmented reality (AR) is poised to evolve into a pervasive technological innovation that permeates both everyday life and professional practice. As a medium that blends digital information with the physical environment, AR has the potential to enhance how individuals perceive and interact with the world around them.
AR can be effectively conceptualized through the reality-virtuality continuum [1], a theoretical framework that delineates the gradual progression from the fully physical world to entirely computer-generated environments. It encompasses both established and emerging technologies that modify sensory experience by integrating or “replacing” elements of the real environment with digitally mediated content. Within this spectrum, four principal categories are typically distinguished: the real environment, augmented reality (in which digital components are superimposed onto the physical world), augmented virtuality (AR) (where virtual environments incorporate selected physical inputs), and fully immersive virtual reality (VR) (which consists solely of digital objects). All these different types of interfaces are nowadays commonly referred to as falling under the umbrella term of Extended Reality (XR).
While VR requires the complete synthesis of an artificial environment, AR maintains direct perceptual contact with the physical world. It enriches the user’s field of view (FoV) with contextually relevant virtual elements anchored in real space. This fundamental distinction affords AR a wide scope of applicability, particularly in contexts where the integration of digital information with ongoing physical activity can enhance the overall quality of human-environment interaction. As such, AR is increasingly considered a promising interface paradigm for sectors ranging from industrial maintenance and design engineering to medicine, logistics, and education.
The rapid evolution of AR also brings forth substantive infrastructural considerations. The potential widespread adoption of this technology will necessitate robust advanced ecosystems capable of sustaining low-latency information flows. This, in turn, will likely intensify demand for edge computing facilities and expanded data center capacity to support real-time rendering pipelines, spatial mapping, object recognition algorithms, and multi-user collaborative environments.
In this article, developed in collaboration with Prof. Thomas Bohné and Dr. Sławomir Tadeja from the Cyber-Human Lab (CHL) at the University of Cambridge, who specialize in development and research (R&D) for immersive systems, we examine augmented reality as a distinct technological domain. We discuss its emerging industrial applications, current hardware exemplified by modern head-mounted displays, prospective developments, practical limitations, and the potential consequent implications for data center expansion.
Although augmented reality remains a comparatively young technological paradigm, its present-day applications already span a broad spectrum of application domains. Today, AR is most commonly associated with either head-mounted displays (HMDs) or mobile devices such as everyday smartphones, which have become the dominant interface for merging digital content with the user’s immediate surroundings.
Popular awareness of augmented reality has been shaped primarily by the entertainment sector, which remains the most recognizable domain of its deployment. Early consumer-grade headsets, such as the PlayStation VR ecosystem, contributed to mainstream acceptance of XR by offering interactive experiences that blurred the boundary between gaming environments and the players’ physical contexts by immersing them in virtual, three-dimensional worlds. The global success of mobile-based Pokémon Go further revealed the cultural and commercial impact of AR by integrating virtual elements into everyday urban space.
In recent years, however, AR has steadily migrated from entertainment into industrial practice, where it increasingly serves as a tool for optimizing workflows. Modern HMDs (exemplified, for instance, by Microsoft HoloLens 2 or Meta Quest 3) support a wide range of applications, including maintenance and diagnostics, guided repair tasks, and real-time monitoring of production processes.
The medical sector has likewise recognized the value of AR as a medium for training, remote care, and patient rehabilitation. AR-based or VR-based simulation environments enable students to rehearse procedures that would otherwise be difficult or risky to replicate. Moreover, AR-supported telerehabilitation programs have proven particularly promising for individuals with chronic conditions.
Educational systems across Europe and elsewhere have begun integrating AR to enhance student engagement. AR allows abstract concepts to be visualized in situ, thereby complementing traditional curricular methods. Museums, archives, and cultural institutions have adopted similar tools to superimpose interpretive layers onto artifacts or historical sites, enriching visitor experience through multimodal storytelling.
AR has reshaped online commerce, particularly in fashion and consumer retail. Virtual try-on systems and product visualization tools enable customers to assess items within their own physical environments, reducing return rates.
Over the past decade, AR devices (particularly HMDs) have undergone substantial transformation in terms of technological sophistication and economic accessibility. Whereas early consumer-oriented headsets routinely exceeded prices of 1000 euros or more, contemporary models are available at a fraction of that cost, with well-known devices now retailing in the range of approx. 340-550 euros (e.g. Meta Quest 3). This decline in price has significantly broadened the potential user base and accelerated the diffusion of AR into everyday and professional contexts.
At the same time, the widespread adoption of AR will likely stem from the availability of mobile devices, the majority of which run Android or iOS. These platforms leverage their massive install base and native software libraries to bring AR to everyday users without the need for dedicated, often high-cost headsets. For developers, this environment is enabled by Google’s ARCore [2] and Apple’s ARKit [3], the proprietary Software Development Kits (SDKs) that provide access to the core capabilities – like motion tracking, plane detection, and light estimation – needed to seamlessly overlay virtual content onto the real world. While ARKit benefits from Apple’s unified hardware ecosystem, offering predictable and high-performance AR experiences, ARCore caters to the vast and diverse Android landscape, with both platforms continuously evolving to enhance spatial understanding and interaction. This mobile-centric approach, fueled by these powerful libraries, ensures that the transformative potential of AR is not limited to specialized devices but is readily available in the pocket of nearly every smartphone user.
A notable early milestone in the trajectory of consumer-facing AR was the introduction of Google Glass in 2013, widely regarded as one of the first commercially available head-mounted AR systems. The device offered functionalities such as identifying individuals in public spaces and synchronizing with social media platforms like Facebook, which quickly provoked public anxiety concerning surveillance and privacy. In certain districts of San Francisco, establishments even displayed signs prohibiting entry to people wearing the device. The ensuing controversy ultimately led Google to withdraw Google Glass from the consumer market. Yet, a decade later, the company is preparing to reintroduce a new generation of AR eyewear [4], expected to be presented publicly next year. In an evolving societal context, where ubiquitous digital technologies have become the norm, this renewed effort reflects a growing openness to wearable AR.
Meta has emerged as one of the most assertive proponents of augmented reality and is often considered to hold a dominant share of the global XR market. In its earlier expansion phase, it was speculated that Meta deliberately priced several generations of its consumer devices (Meta Quest 2 & Meta Quest 3) below production cost in order to stimulate market adoption [5]. Although this strategy seemed to have been eventually suspended, the company has allocated extraordinary financial resources that can be counted in billions of USD annually – spent on Reality Labs, Meta’s AR and VR research arm [6].
The global trajectory of augmented reality is increasingly shaped by two principal actors: the United States and China. The latter has embedded immersive technologies within its broader national strategy for digital modernization, supported by a state-driven expansion of data infrastructure. One of the key institutional actors in this landscape is the Industry of Virtual Reality Alliance (IVRA), which seems to play one of the vital roles in accelerating the commercialization of VR and AR technologies in China. Meanwhile, the United States, despite being home to many of the field’s most influential companies and research laboratories, faces a noticeable infrastructural impediment, as uneven network coverage across large regions limits the scalable deployment of AR. This is particularly consequential given that AR ecosystems, especially in the context of using mobile devices to deliver immersive experiences, depend on communication networks capable of sustaining high-volume, low-latency data transmissions; conditions typically associated with dense 5G, and eventually 6G, infrastructure.
Europe, on the other hand, has adopted a more cautious approach, with AR uptake remaining concentrated primarily in consumer entertainment and gaming.
Contemporary AR systems continue to face substantial technological constraints that hinder their broader adoption. Beyond the infrastructural requirements already noted (the dependence on advanced 5G networks), the principal constraint remains the physical configuration of existing devices. Current HMDs are often sizable and are weighty, factors that substantially limit their suitability for prolonged use. Apple’s recent headset, for instance, comes with the producer’s recommendation that users take breaks after 20 to 30 minutes of use when becoming accustomed to the device. The device’s underwhelming market performance ultimately prompted Apple to scale-back production. In the longer term, continued miniaturization may yield genuinely lightweight wearables.
A further technical challenge concerns the restricted field of view. Whereas human peripheral vision encompasses roughly 170 degrees, most AR headsets provide only a narrow segment of this range, thereby constraining the immersiveness of the interface.
Until recently, another significant obstacle was the persistent shortage of high-quality digital content designed specifically for AR environments. However, recent advances in generative artificial intelligence (AI) have begun to mitigate this limitation by enabling the more efficient creation and scaling of visual, auditory, and interactive materials.
While it is plausible that many of these limitations will be progressively alleviated over the coming years, the dynamics of societal acceptance and market diffusion constitute distinct challenges. Despite a marked increase in user receptivity relative to the earlier period of Google Glass, it is probable that manufacturers may need to subsidize devices, potentially pricing them at their production cost, to establish a sufficiently large consumer base. At the same time, initiatives like Meta’s partnership with the producer of Ray-Ban glasses represent a strategic move integrating sophisticated yet limited AI-powered features (e.g. voice commands, calls, music, and real-time information) into a familiar, fashionable form factor [7]. Their accessible pricing, combined with the desirability of the Ray-Ban brand, helps broaden the consumer base and get people accustomed to wearing connected devices on their faces.
The prospective trajectory of augmented reality invites a comparison with an earlier technological transition. Just as smartphones gradually subsumed many of the core functions once reserved for personal computers, some experts speculate that lightweight AR headsets may, in time, assume a portion of the roles currently fulfilled by mobile phones. Whether such a shift will materialise remains contingent upon the emergence of devices that are socially acceptable and technically advanced. This objective may become attainable within the next several years should ongoing efforts in hardware miniaturisation succeed in producing ergonomically viable headsets.
If such devices were to gain widespread acceptance, their price is likely to be comparable to that of contemporary flagship smartphones. AR’s functional potential would be further amplified by integration with AI personal agents capable of furnishing context-sensitive guidance in real time. One might imagine, for instance, navigational visual and audio cues such as maps appearing directly within the user’s visual field, or real-time annotations identifying buildings, colleagues, or points of interest as the user moves through a given environment.
Should these developments converge (advances in hardware, accessible pricing, and real-time augmentation), AR could evolve into a pervasive interface layer mediating daily interaction with digital systems and the physical world.
The diffusion of augmented reality is poised to exert an additional stress on data centers. Should hyperscale providers (including companies such as Meta, whose strategy has increasingly prioritized AR/VR ecosystems and their coupling with AI) continue along this trajectory, a logical consequence will be the progressive densification of edge-computing facilities within metropolitan areas. AR applications, particularly those involving real-time spatial mapping, object recognition, and multi-user interaction, are sensitive to latency. To maintain the illusion of seamless perceptual continuity, computational processes should be able to outpace the human brain’s ability to detect noticeable delays. This requirement may need to be met by locating compute resources in close proximity to their end-users.
The energy footprint associated with AR workloads is likewise expected to expand. This is due both to the computational demands of rendering and sensor fusion inherent in XR systems, and to the incorporation of generative artificial intelligence for the production of content in AR applications. As energy consumption rises, data centers will require enhanced power provisioning and advanced cooling solutions to sustain high-intensity workloads.
In the context of emerging smart cities, augmented reality is increasingly positioned as a vital mediating layer between complex, sensor-driven urban data ecosystems and the lived experience of their residents. As municipalities move toward deploying interconnected Internet of Things (IoT) networks, autonomous mobility systems, and digitally integrated public services, AR interfaces are likely to become one of the principal modalities through which individuals perceive and engage with this continuous flow of information [8]. Sustaining such dynamic, context-aware overlays (whether for navigation, safety notifications, or access to public services) requires low-latency computational support. This dependency underscores the strategic importance of metropolitan edge data centers, which must swiftly process and synthesize urban datasets to maintain real-time AR functionality. With the ongoing expansion of smart-city infrastructure, the interdependence between AR-enabled civic services and distributed computing will intensify, effectively positioning edge data centers as indispensable components of the urban digital fabric.
[1] Milgram, Paul; H. Takemura; A. Utsumi; F. Kishino (1994). “Augmented Reality: A class of displays on the reality-virtuality continuum”. Proceedings of SPIE – The International Society for Optical Engineering Vol. 2351
[2] https://developers.google.com/ar
[3] https://developer.apple.com/augmented-reality/arkit/
[4] https://www.bbc.com/news/articles/cwyx83n00k6o
[5] https://tech.yahoo.com/ar-vr/articles/much-meta-quest-3-really-200333825.html
[6] https://virtual.reality.news/news/meta-slashes-metaverse-budget-30-after-70b-loss/
[7] https://www.ray-ban.com/usa/ray-ban-meta-ai-glasses; https://www.meta.com/ai-glasses/ray-ban-meta/
[8] Sharifi, A.; Amirzadeh, M.; Khavarian-Garmsir, A.R. (2025). “The metaverse as a future form of smart cities: A systematic literature review of co-benefits and trade-offs for sustainable development goals”. Cities, vol 161, 105879, ISSN 0264-2751, doi: 10.1016/j.cities.2025.105879
Prof. Thomas Bohné,
University of Cambridge
Dr. Sławomir Tadeja,
University of Cambridge
This brochure presents a comprehensive overview of the data center sector as a cornerstone of the global digital economy, driven by exponential growth in AI, cloud computing, and data flows. It examines Warsaw’s emergence as a key Tier 2 market in Europe, supported by advanced digital infrastructure, competitive operating conditions, and strategic geographic positioning. The document addresses the critical role of energy access and sustainability, with a focus on renewable integration, power resilience, and heat recovery.
You can now download the report: