On September 15, 2026, Sharp Corporation announced its entry into the domestic AI server market under its own brand starting in fiscal 2027, in partnership with parent company Foxconn (Hon Hai Precision Industry), targeting ¥250 billion in annual revenue by fiscal 2030. Currently undergoing structural reform, Sharp aims to reignite growth by capturing Japan’s rapidly expanding demand for generative AI infrastructure.

The Full Picture of Sharp’s AI Server Entry Targeting ¥250 Billion by FY2030

On September 15, 2026, Sharp Corporation held a press conference in Tokyo to officially announce its entry into the AI infrastructure sector, where demand is surging due to the rapid spread of generative AI. Tetsuji Kawamura, Representative Director, President, Executive Officer and CEO, emphasized that this initiative is not merely a new business venture, but a vital pillar driving the company’s broader transformation. Key details of the announcement include:

• Launch Schedule: Full-scale commercial sales under the Sharp brand in Japan starting in fiscal year 2027
• Order Commencement: Orders for enterprise-tier models opened on September 15, 2026
• Revenue Target: Achieving ¥250 billion in annual sales by fiscal year 2030
• Target Customers: Enterprises, local governments, research institutions, and data center operators

Sharp has been pursuing structural reform following downturns in its LCD panel business. By combining its brand recognition and domestic sales channels with the advanced manufacturing capabilities of its parent company, Taiwan-based Foxconn (Hon Hai Precision Industry), Sharp aims to achieve early monetization. As shown in the diagram below, the company plans to rapidly establish a robust AI server supply framework to serve as essential infrastructure

.

Performance Comparison of NVIDIA GPU-Powered Models: Air-Cooled vs. Liquid-Cooled

Sharp’s AI server lineup comprises two models, both equipped with the latest GPUs from US semiconductor giant NVIDIA. The company offers solutions scaled to meet diverse operational needs, ranging from edge environments to large-scale data center infrastructure.

• Enterprise AI Server (NVIDIA RTX Pro Server / Model: AI-B43100): A 4U rack-mount server capable of housing up to eight high-performance NVIDIA RTX PRO 6000 Blackwell Server Edition GPUs. Powered by dual Intel Xeon 6 series processors and up to 3TB of DDR5 memory, it utilizes an air-cooling system for seamless deployment into existing enterprise facilities. Delivering an inference performance of 32 PFLOPS, orders officially opened on September 15, 2026.
• Large-Scale Training and Inference AI Server (NVIDIA HGX System): A high-end system featuring the NVIDIA HGX Rubin NVL8, which interconnects eight GPUs via NVIDIA NVLink. Designed for training and inference of high-density large language models (LLMs), it employs a liquid-cooling architecture, with orders scheduled to begin in fiscal year 2027.

Synergies with Foxconn Group and Domestic Supply Infrastructure

Structural Pivot from LCD Panels to Asset-Light Management

The catalyst behind Sharp’s entry into the AI server market lies in its ongoing business restructuring and transition toward an asset-light operating model. In August 2024, the company halted large LCD panel production at its subsidiary, Sakai Display Products (SDP), marking a complete exit from the television LCD panel manufacturing business. The land and facilities at the Sakai plant were subsequently transferred to SoftBank and KDDI, where they are currently being repurposed into large-scale AI data centers.

As part of its shift away from capital-intensive device businesses that generated substantial losses toward brand-focused and high-value B2B solution businesses, Sharp executed a Memorandum of Understanding (MoU) for strategic collaboration on new businesses with parent company Foxconn on June 24, 2026. Aligned with Foxconn’s “3+3+3 Strategy” (EVs, digital health, and robotics × AI, semiconductors, and next-generation communications), Sharp is establishing a robust collaborative business development platform, as illustrated below

.

Combining Foxconn’s Manufacturing Scale with Domestic Sales and Maintenance Networks

In manufacturing and component procurement, Sharp will fully leverage Foxconn Group’s global supply chain and manufacturing scale—which commands approximately 40% of the global AI server market—amid ongoing global supply tightness in semiconductors and memory. Initially, Sharp will procure servers manufactured at Foxconn facilities, conduct quality assurance checks in Japan, and distribute them locally as proprietary Sharp-branded products.

To accelerate market penetration across Japan, Sharp has established a comprehensive alliance network:

• Sales Network: Basic agreements have been signed with two leading Japanese IT distributors, Daiwabo Information System and Ryosan Ryoyo, to utilize their nationwide sales channels.
• Operations & Maintenance: Partnered with Tomorrow Net to deliver end-to-end support ranging from system deployment to ongoing maintenance.

Furthermore, Sharp is considering establishing local manufacturing in Japan over the long term. CEO Kawamura revealed that the company is actively evaluating options, designating its Kameyama Plant in Mie Prefecture as a prime candidate for domestic server production.

Shifting Market Dynamics and Long-Term Growth Outlook

The Impact of a ¥7 Trillion Domestic Market Driven by Sovereign AI and Edge Computing

According to market forecasts released by Sharp, Japan’s domestic AI server market is projected to expand from approximately ¥0.9 trillion (¥900 billion) in FY2025 to ¥4 trillion in FY2030, eventually reaching ¥7 trillion by FY2035. This represents a rapid compound annual growth rate (CAGR) of 22.7%, propelled by structural transformations across the technology landscape.

Key drivers accelerating this demand include:

• Strategic Shift from Training to Inference: As AI adoption matures, server demand dedicated to routine inference workloads continues to compound.
• Expansion of Physical AI and Robotics: The rise of factory automation and industrial robotics requires real-time processing, driving exponential increases in data volume.
• Rise of Sovereign AI and On-Premises Migration: Heightened emphasis on data sovereignty, strict confidentiality, and power grid constraints in urban centers are pushing organizations away from total cloud dependency toward hybrid architectures combining on-premises and edge AI processing.

Market Competition, Differentiation Strategy, and Future Milestones

Japan’s rapidly expanding AI server market is already a fiercely contested arena featuring global giants such as Dell Technologies, Hewlett Packard Enterprise (HPE), and Super Micro Computer (SMCI), alongside domestic IT vendors. Sharp’s path to winning market share rests on its unique value proposition: uniting Foxconn’s cost efficiency and component procurement power with Sharp’s high-touch domestic sales and maintenance network to offer superior price competitiveness and reliability.

On the day of the announcement (September 15, 2026), Sharp’s stock slipped in Tokyo trading, snapping a four-day winning streak. Analysts attributed the modest market reaction to the medium-term nature of the FY2030 revenue target (¥250 billion) and lingering questions regarding early profit margins. Moving forward, key catalysts to monitor include the opening of orders for the NVIDIA HGX System in FY2027, finalized decisions regarding domestic production at the Kameyama Plant, and the creation of new synergies between Sharp’s AI server business and its existing IoT appliance and enterprise B2B solution portfolios.

Reference Pages

• [Press Release] Entry into the AI Server Business (Sharp Corporation) https://corporate.jp.sharp/news/260915-a.html
• [Press Release] Signing of Memorandum of Understanding on Strategic Collaboration in New Businesses (Sharp Corporation & Hon Hai Precision Industry Co., Ltd.) https://corporate.jp.sharp/news/260624-a.html

【#Sharp #AIServer #Foxconn #NVIDIA #SovereignAI #DataCenter #KameyamaPlant #ITBusiness】