Unlike the early 5G era, 6G is beginning to take shape as an AI-native generation of networks. The conversation is no longer defined simply by faster speeds or greater device density, but is moving to building systems that can sense, reason, adapt, and act in real time across highly distributed environments.
Rather than being confined to a central core, intelligence is orchestrated across endpoints, edge nodes, private infrastructure, and the cloud. This shift positions 6G as a foundational layer for physical AI, autonomous mobility, industrial automation, smart environments, and other latency-sensitive digital systems.
FPT to Enable the AI-Native Leap
The AI-native character of 6G lies not in using AI as an optimization tool, but in embedding intelligence into the fabric of the network itself. Viewed through that lens, AI-native 6G is not only a telecom story, but also one of infrastructure and systems integration. It requires industrial-scale AI infrastructure, robust data pipelines, model lifecycle management, and the ability to run continuous training and inference across distributed environments. Across the market, that readiness is increasingly defined by the ability to industrialize AI end to end. The stack, therefore, is not built as an isolated research infrastructure, but as a foundation for taking AI from development into enterprise-scale deployment with the underlying AI environment supporting not only performance, but also security, privacy, and operational control.
FPT AI Factories, powered by NVIDIA H100 and H200 GPUs, reflects that broader shift, bringing together high-performance compute, data pipelines, model development and testing environments, production serving capabilities, and a marketplace layer for reusable AI components. Built with operational maturity in mind, the platform is further reinforced by SOC 2 and SOC 3 certifications that speak to the governance and assurance increasingly expected of large-scale AI workloads and datasets.
That foundation is supported by workforce scale and long-term capability building. FPT now boasts over 25,000 AI-augmented engineers applying AI across real-world delivery, alongside a growing pipeline of more than 2,000 AI and Data Engineering graduates annually from FPT University. AI education is also being expanded across FPT’s broader education system, with early exposure to AI and robotics from the primary level; while research exchanges, residency programs, and PhD pathways connect Vietnamese talent to global research networks and deepen future-ready capabilities across AI, chip design, and computational engineering.
At the platform layer, FPT’s AI-first platform FleziPT extends that foundation into execution. It reflects the company’s approach to embedding AI into software engineering, operations, and large-scale platforms, rather than treating it as a conceptual layer added after deployment. Through FleziPT, FPT addresses these demands through an AI-driven software development life cycle, horizontal AI agents, vertical accelerators, and a secure-by-design architecture shaped by governance and responsible AI principles. The emphasis is on faster iteration, repeatable deployment, and stronger control across complex environments which delivers up to 50% faster deployment rollouts, around 40% lower operational overhead, and twice the speed from idea to MVP.
These capabilities are also recognised by independent market assessments, notably as a Leader in IDC MarketScape Asia Pacific AI-Enabled Front Office Conversational AI Software 2025 and a Major Contender in the Everest Group Application Development Services for AI Applications PEAK Matrix® Assessment 2025. It is already evident in live deployments across the distributed, data-intensive environments aligned with the demands of AI-native 6G. In enterprise operations, FPT developed IvyAgents as a multi-agent AI assistant to automate data access, document processing, translation, and customer support at scale. The solution helped a leading trade company reduce processing time by up to 90%, lower human costs by 33%, and cut error rates by 80%. In manufacturing, FPT built an on-premise conversational GenAI system for a Japanese manufacturer, enabling multilingual access to production and operational data across MES, ERP, and SQL-based systems. The platform reduced information retrieval time by 60% while improving visibility across factory layers.
This same complexity also explains why AI-native 6G cannot be treated as a single-vendor ambition. Success will hinge on how effectively the ecosystem can align silicon, software, cloud platforms, edge infrastructure, network architecture, standards, and deployment models. Early coalitions help reduce that risk by giving the ecosystem a shared roadmap, common validation points, and a clearer path toward commercialization. They offer a way to compress learning cycles, lower time-to-market risk, and begin building interoperability before the market reaches maturity.
Qualcomm’s recent coalition reflects a broader industry recognition that AI-native 6G will be shaped not only by breakthrough technologies, but by how quickly those technologies can be tested, aligned, and translated into deployable capabilities. Participating in the alliance, FPT brings strengths in large-scale digital engineering, AI, and system integration to the practical challenge of turning early AI-native 6G architecture into deployable capability. Those capabilities are also increasingly relevant across the domains expected to benefit most from AI-native 6G, including intelligent mobility, autonomous UAV systems, smart manufacturing, and other advanced digital engineering environments.
Prior to the emergence of 6G, FPT has been one of Ericsson’s first 5G partners in Vietnam, co-developing and deploying innovative use cases for both consumer and enterprise sectors. The tech firm also established strategic collaborations with major tech leaders such as NVIDIA, Microsoft, and SAP strengthen its access to core AI, cloud, and enterprise capabilities, while sovereign AI partnerships with organizations like Sumitomo and SBI Holdings in Japan point to another dimension of readiness—the ability to adapt AI development to market-specific requirements, governance expectations, and operational contexts. Combined with partnerships that expand AI training and development, including those with NVIDIA, LandingAI, and Mila Quebec AI Institute, this ecosystem does more than broaden technical capability. It helps build the engineering discipline, decision-making maturity, and market alignment needed to move emerging AI systems from ambition to adoption.
Toward a More Intelligent Network Future
The next generation of networks is being shaped not only by faster transmission, but by the integration of intelligence into the fabric of the system itself. That future will depend on ecosystem coordination, software maturity, and the ability to operationalize AI across highly distributed environments. Leveraging its experience across digital engineering, AI, and large-scale systems integration, FPT is well positioned to translate the promise of AI-native 6G into the real-world capabilities future industries will require.
