Tareq Amin in Saudi Arabia’s AI Compute Landscape
Tareq Amin represents a category of participant in Saudi Arabia’s AI compute ecosystem that is distinct from the corporate entities, government agencies, and sovereign investment vehicles that dominate the landscape: the internationally prominent Saudi-origin technology executive whose ideas, advocacy, and institutional relationships actively shape the direction of Saudi digital transformation. As CEO of Rakuten Mobile in Japan and one of the global architects of Open RAN telecommunications, Amin carries a profile that commands attention at the intersection of Saudi AI ambitions and 5G infrastructure modernization — two domains that are more deeply connected than they might initially appear.
Born in Saudi Arabia and educated in the Kingdom before building his career in international telecommunications, Amin embodies a particular strand of the Saudi diaspora tech leadership that Vision 2030’s architects are actively trying to engage and, in some cases, repatriate. His presence on the main stage at LEAP — the annual Saudi tech conference organized by the Communications and Information Technology Commission — and at FII (the Future Investment Initiative, organized by PIF) places him in direct dialogue with the Saudi decision-makers driving the Kingdom’s AI infrastructure investments. He is not merely a speaker at these events; he is a substantive interlocutor whose technical authority on network architecture and AI-at-the-edge positions him as an advisor to Saudi telecoms modernization programs.
Amin’s most significant technical contribution — the leadership of Rakuten Mobile’s buildout of Japan’s first cloud-native, fully virtualized 4G/5G network — is directly relevant to Saudi Arabia’s AI compute ambitions because it demonstrated, at commercial scale, that a disaggregated Open RAN architecture could support the kind of low-latency, high-density connectivity that AI inference at the network edge requires. Saudi Arabia’s 5G infrastructure is being constructed with AI deployment in mind from the outset, and the architectural choices made in that construction — centralized versus distributed radio access, proprietary versus open interfaces, monolithic versus cloud-native core networks — will determine whether Saudi 5G can serve as a delivery mechanism for AI services at the national scale that Vision 2030 envisions.
Amin’s role is therefore simultaneously technical, advisory, and symbolic: technical in the substance of what he advocates, advisory in his direct engagement with Saudi telecoms and AI policy, and symbolic in what his prominence represents for Saudi ambitions to be recognized as a global technology leader rather than a technology consumer.
Strategic Significance
The strategic significance of Tareq Amin to Saudi Arabia’s AI compute ecosystem is concentrated in two areas: his advocacy for Open RAN as the architectural foundation for 5G AI delivery infrastructure, and his embodiment of the Saudi diaspora tech talent that Vision 2030 is attempting to leverage.
Open RAN’s relevance to AI compute is architectural. Traditional cellular networks are built on proprietary, vertically integrated hardware from vendors like Ericsson, Nokia, and Huawei, where radio hardware, baseband processing, and network management software are tightly coupled and delivered as integrated systems. Open RAN disaggregates this stack: the radio unit (RU), distributed unit (DU), and centralized unit (CU) are separated by standardized interfaces, allowing operators to mix hardware from different vendors and run baseband processing on commercial off-the-shelf servers in standard data centers. This disaggregation has a direct consequence for AI deployment: when baseband processing runs on commercial servers in standard data centers, those same data center resources can also run AI inference workloads during periods of low network traffic, or AI can be co-located with radio processing to analyze network traffic patterns and optimize resource allocation in real time.
Amin’s work at Rakuten Mobile proved this architecture at scale. Rakuten built Japan’s fourth mobile network from scratch on an O-RAN compliant, cloud-native foundation, running the entire network core on Rakuten’s own cloud infrastructure and using commercial server hardware for radio signal processing. The network now serves millions of subscribers and has demonstrated that Open RAN can deliver carrier-grade reliability while enabling a degree of software-defined flexibility that proprietary networks cannot match. For Saudi telecoms operators — stc, Mobily, and Zain — who are in the midst of 5G network expansion and are evaluating Open RAN adoption as part of that expansion, Rakuten’s network is the most credible proof point available that Open RAN works at national scale.
The direct relevance to Saudi AI deployment is through the edge AI inference architecture. AI inference at the network edge — running AI models close to end users to minimize latency, rather than routing inference requests to centralized cloud data centers — requires compute resources distributed across the radio access network. In an Open RAN architecture, the distributed units (DUs) that handle radio signal processing sit in local data centers co-located with cell towers, and those same facilities can host AI inference accelerators. Qualcomm’s Cloud AI 100 Ultra and Intel’s Gaudi 3 are both suitable for this edge deployment pattern. For Saudi Arabia’s NEOM project — which is explicitly designed as a technology-forward city where AI-powered services will be delivered across a dense 5G network — the edge AI architecture that Open RAN enables is not a future possibility but a design requirement.
Amin’s advisory engagement with Saudi telecoms modernization encompasses this architectural question directly. His recommendations to Saudi operators and to CITC — the Communications, Space and Technology Commission that regulates Saudi telecoms — have consistently pushed toward Open RAN adoption as part of 5G buildout, both for the operational flexibility it provides and for the supply chain diversification it enables. Given that Huawei’s participation in Saudi 5G infrastructure is subject to ongoing geopolitical sensitivity, Open RAN’s vendor-neutral architecture provides Saudi operators with a pathway to reducing Huawei dependency without being forced to consolidate on Ericsson or Nokia.
Operational Context
Amin’s operational engagement with Saudi Arabia occurs across several channels: his appearances at Saudi-hosted conferences, his formal advisory relationships with Saudi telecoms entities, and the ongoing evaluation of Rakuten’s O-RAN platform by Saudi 5G operators.
At LEAP and FII, Amin is a recurring presence whose talks consistently address the convergence of 5G infrastructure and AI deployment. His 2024 LEAP keynote addressed the specific question of how Open RAN architectures enable AI inference at the network edge, framing it as a critical infrastructure choice for countries building AI-enabled societies from the ground up — a framing directly applicable to Saudi Arabia’s situation. These appearances are not mere marketing; they are substantive technical presentations that reach the MCIT officials, telecoms executives, and AI researchers who make Saudi infrastructure decisions.
The evaluation of Rakuten’s O-RAN platform by Saudi 5G operators is an active commercial process. Both stc and Mobily have conducted technical assessments of Open RAN solutions as part of their 5G RAN refresh programs, and Rakuten Symphony — the Rakuten subsidiary that commercializes Rakuten’s O-RAN technology stack for global operators — has been involved in those assessments. Amin’s personal engagement with Saudi telecoms executives, facilitated by his LEAP and FII presence, has accelerated these evaluations by providing executive-level credibility for what might otherwise be dismissed as an immature technology.
The practical deployment question centers on specific network segments. Saudi operators are unlikely to rip and replace existing 4G infrastructure with Open RAN equivalents, but new 5G radio deployments — particularly in greenfield areas like NEOM and the Red Sea development project, and in the dense urban areas of Riyadh being upgraded for AI-era connectivity — are realistic candidates for Open RAN deployment. Amin’s operational insight, derived from Rakuten’s commercial deployment experience, is directly applicable to these Saudi deployment scenarios: Rakuten faced similar greenfield versus brownfield trade-offs in building Japan’s fourth network from scratch alongside three established incumbents.
Amin’s engagement with Saudi AI compute infrastructure extends beyond telecoms to the broader question of cloud-native architecture. His advocacy for cloud-native network functions — software-defined network elements that run on commercial cloud infrastructure rather than proprietary hardware — aligns with the direction that Saudi AI compute operators are taking as they build infrastructure intended to host both AI training workloads and network functions in a shared, software-defined environment. The convergence of AI compute and network function virtualization, both running on the same pool of commercial server hardware in standard data centers, is a direction that Amin has advocated for internationally and that Saudi infrastructure planners are taking seriously.
Connections to the Broader Saudi Ecosystem
Amin’s connections to the Saudi AI compute ecosystem are primarily relational and advisory rather than transactional, which gives them a different character but not less importance than the commercial relationships that define most ecosystem participants.
The most direct connection is through the Saudi telecoms operators — stc, Mobily, and Zain — who are Rakuten Symphony’s prospective customers for O-RAN technology. stc in particular, as Saudi Arabia’s incumbent operator and a PIF portfolio asset with a strategic mandate to support Vision 2030’s digital infrastructure goals, is the most important potential adopter of Open RAN in the Kingdom. Amin’s relationship with stc’s leadership, cultivated through multiple years of conference interaction and direct advisory engagement, positions Rakuten Symphony as a preferred consideration in stc’s Open RAN evaluation process.
The connection to NEOM is substantive. NEOM’s technology infrastructure team has engaged with Amin and Rakuten’s O-RAN experts in evaluating the communications architecture for The Line and the broader NEOM project zone. NEOM’s requirement for ultra-dense 5G coverage with edge AI inference capabilities maps closely onto the Open RAN architecture that Rakuten has deployed in Japan. The question of whether NEOM adopts a proprietary RAN architecture from Ericsson or Nokia, or pursues an Open RAN approach that would allow greater flexibility in integrating AI compute resources with radio infrastructure, is a decision with significant long-term implications for the project’s AI service delivery capabilities.
Humain’s AI infrastructure agenda creates a connection through the edge AI inference dimension. Humain is not merely building centralized GPU clusters; its mandate includes enabling Saudi AI capabilities to be delivered across the Kingdom’s geography, including in secondary cities and rural areas. Achieving that geographic reach at acceptable latency requires either a massive expansion of centralized compute with high-capacity network connectivity, or a distributed edge AI architecture where inference compute is co-located with radio infrastructure. The latter architecture is exactly what Amin advocates, creating alignment between Humain’s deployment challenge and Rakuten’s technical approach.
SDAIA and MCIT both engage with Amin through the conference circuit and through formal consultations on Saudi digital infrastructure strategy. His advice on the interplay between network architecture and AI service delivery has informed MCIT’s thinking on 5G deployment requirements, and SDAIA’s National AI Strategy includes provisions for edge AI deployment that reflect the architectural thinking Amin has advocated.
Outlook: 3–5 Year Trajectory
Amin’s trajectory in relation to Saudi Arabia’s AI compute ecosystem over the 2025–2030 period will be shaped by two variables: whether Open RAN achieves significant deployment scale in Saudi 5G networks, and whether Saudi Arabia succeeds in attracting diaspora tech leadership back to the Kingdom in operational roles rather than merely advisory ones.
On Open RAN adoption, the Saudi 5G market is at an inflection point. The initial 5G deployments by stc, Mobily, and Zain in 2022–2024 used predominantly proprietary RAN equipment from established vendors, but the 5G network densification phase that is now beginning — adding thousands of small cells in urban areas and extending coverage to industrial zones, logistics hubs, and smart city deployments — is where Open RAN’s economic advantages and architectural flexibility become most compelling. If stc or Mobily makes a significant Open RAN commitment in this densification phase, Amin’s advisory role will convert into a commercially significant relationship between Rakuten Symphony and Saudi telecoms operators.
The question of whether Amin himself plays a more direct role in Saudi AI infrastructure — through a formal advisory board position with Humain, a consulting engagement with NEOM’s technology team, or potentially through a return to the Kingdom in an operational role — is the most consequential uncertainty. Vision 2030’s talent repatriation agenda specifically targets Saudi professionals of Amin’s caliber, and the government has created frameworks — including citizenship incentives, housing programs, and leadership positions in Vision 2030 delivery entities — to make return attractive. Whether Amin ultimately engages with Saudi AI infrastructure in an advisory or operational capacity, his intellectual influence on Saudi 5G and edge AI architecture decisions is already embedded in the policy discussions that will shape the Kingdom’s infrastructure for decades.