Special Feature: 2026 International Consumer Electronics Show (CES) During CES 2026, Ashley Gorakhpurwalla, President of Lenovo's Infrastructure Solutions Group (ISG), systematically detailed the strategic partnership between Lenovo and NVIDIA in response to questions. He emphasized that their collaboration is not merely a single-product-level tie-up but a full lifecycle synergy covering planning, manufacturing, deployment, and operations. Ashley Gorakhpurwalla stated that understanding the Lenovo-NVIDIA relationship first requires recognizing the high flexibility of Lenovo ISG's business model. Lenovo can operate as an ODM, where partners provide core technology while Lenovo handles manufacturing, deployment, support, and full lifecycle management. Simultaneously, Lenovo is continuously introducing its own IP and possesses a mature portfolio of products supported globally across 180 countries. This enables Lenovo to serve hyperscale clients, neocloud providers, traditional service providers, and even enterprise clients needing just a single server or storage device, covering the entire spectrum from top-tier cloud to enterprise IT. Building upon this complete spectrum, Lenovo and NVIDIA have established a multi-layered, long-term cooperative relationship. Ashley Gorakhpurwalla recalled that during the keynote speech, NVIDIA founder and CEO Jensen Huang and Lenovo Chairman and CEO Yang Yuanqing jointly highlighted several of Lenovo's core advantages in the AI infrastructure domain. First, Lenovo possesses manufacturing capabilities at a global scale, with existing capacity, power resources, and delivery capabilities to support the deployment of large-scale AI systems. Second, both during the Lenovo era and tracing back to the IBM days, Lenovo has accumulated decades of experience in building large-scale systems, encompassing mainframes, high-performance computing, and supercomputers. Such systems typically consist of numerous racks operating as a unified whole within data centers, and Lenovo has developed mature capabilities in integrating and delivering these complex systems. Third, and particularly crucial for current AI training systems, is Lenovo's ownership of the core IP for Neptune liquid-cooling technology. Ashley Gorakhpurwalla pointed out that in large-scale AI training scenarios, liquid cooling is becoming the only sustainable deployment method. Without significantly reducing overall data center energy consumption through liquid cooling, global power resources would struggle to support the continuous growth of AI training demands, an area where Lenovo has long focused and holds a leading advantage. Leveraging these capabilities, Lenovo and NVIDIA are jointly providing a replicable "blueprint" for global AI training clouds, inference clouds, and new cloud architectures. This helps clients accelerate the transition from funding and conception to actual business implementation. Ashley Gorakhpurwalla stated that Lenovo does not directly build data centers itself, but after a client completes planning, Lenovo provides consulting, technical support, and equipment deployment to ensure the data center can quickly progress from infrastructure construction to actual output. Concurrently, Lenovo will also manufacture, deploy, and support NVIDIA's related capabilities based on the Rubin architecture on NVIDIA's behalf. Regarding the enterprise market, Ashley Gorakhpurwalla indicated that most enterprise clients are not aiming to train large models but rather want to use AI to enhance business efficiency, a goal consistent with Lenovo's own internal practices. At this level, Lenovo not only cooperates with NVIDIA but also maintains partnerships with AMD and other accelerator manufacturers to help enterprises deploy the most suitable technology solutions in the cloud, hybrid environments, or at the edge. He noted that this multi-dimensional collaboration is not limited to servers or data centers but extends to multiple business segments including workstations and client devices, giving the Lenovo-NVIDIA partnership greater systematic and strategic depth. Building on this foundation, Arthur Hu, Senior Vice President of Lenovo Group, Chief Information Officer, and Chief Technology & Delivery Officer of SSG, provided additional commentary on the "last mile" problem in AI implementation. He pointed out that as Agentic AI gradually became the mainstream way for consumers and enterprises to use AI in 2025, this trend is expected to strengthen further in 2026. The key to AI truly creating value lies in how quickly it can enter a sustainable "consumption" phase. Arthur Hu stated that Lenovo is collaborating with NVIDIA through an agent platform to help enterprises accelerate the transition from deployment to use. With the launch of the xIQ Agent platform at CES, Lenovo aims to continue compressing cycles and improving efficiency in areas such as data management, continuous operations, agent building, evaluation, and optimization after hardware delivery is complete. The goal is to help clients achieve the "first token" with tangible business significance more rapidly.
Comments