Why Toilets and MSG Are Winning the AI Boom | Bloomberg Tech: Asia 6/26/2026

Key Concepts

• AI Supply Chain: The multi-layered infrastructure (power, chips, packaging, cooling) required to scale AI from proof-of-concept to real-world economic impact.
• Hardware-Centric AI: The perspective that AI is not just a software story but a physical, hardware-intensive industry where Asia holds a strategic advantage.
• Bottlenecks: Critical supply chain constraints including memory (DRAM/HBM), ABF substrates, raw silicon wafers, and skilled labor.
• Sovereign AI: Government-led initiatives (notably in Asia) to develop domestic AI models and infrastructure to boost national productivity.
• Corporate Governance Reform: A trend in Japan where companies are becoming more responsive to shareholder engagement and focusing on valuation.
• Ancillary AI Plays: Established industrial companies (e.g., ceramics, seasoning makers) that possess critical, overlooked components for AI infrastructure.

---

1. The Shift to Hardware-Centric AI in Asia

The global AI boom is driving a massive capital flow into Asia, which serves as the backbone of the AI supply chain. While the West (specifically the US) focuses on software models and hyperscale platforms, Asia dominates the "five-layer AI cake"—power, chips, infrastructure, models, and applications.

• Capital Efficiency: Unlike the US, where hyperscalers are self-funding or tapping debt markets, Asian AI development is often supported by a mix of sovereign wealth funds, public pensions, and private infrastructure funds.
• Productivity Focus: Asian economies are leveraging AI to improve productivity within existing infrastructure, rather than solely focusing on massive new build-outs.

2. Critical Bottlenecks and Supply Chain Dynamics

Experts from SemiAnalysis highlight that as demand for AI servers surges, the supply chain is tightening across several specialized sectors:

• Memory (DRAM/HBM): Identified as the most difficult component to ramp up due to manufacturing complexity and long lead times for building new fabs.
• ABF Substrates: Essential for chip packaging; dominated by companies like Ibiden, Shinko, Unimicron, and Nanya. Increased chip size and complexity are causing severe supply tightness.
• Silicon Wafers: Leading-edge logic manufacturing relies on epitaxial silicon wafers, a market dominated by Japanese firms Shin-Etsu and Sumco.
• Manufacturing Constraints: The industry faces a "triple threat" of limited clean room capacity, specialized equipment (e.g., ASML lithography scanners), and a shortage of skilled engineering talent.

3. Unexpected AI Beneficiaries: Case Studies

Investors are moving upstream to identify "choke points" in the supply chain, leading to interest in companies previously unrelated to the tech sector:

• Toto: Traditionally a toilet manufacturer, its advanced ceramics division is now a critical beneficiary of AI memory production, as its ceramics can withstand the high temperatures and corrosive materials used in chip manufacturing.
• Ajinomoto: Known for MSG, the company holds a 90%+ global market share in Ajinomoto Build-up Film (ABF), an essential insulating material for data center servers. They have significant pricing power due to the lack of viable alternatives.
• WUS (Taiwan): A leader in the PCB (Printed Circuit Board) stack for AI servers, with a 45% market share in its specific niche.

4. Strategic Perspectives and Market Arguments

• The "Choke Point" Strategy: Palliser Capital argues that the best investment opportunities lie in companies that control essential, high-barrier-to-entry components. These firms have deep "moats" and significant runway for price increases.
• China’s Sovereign AI: China is aggressively pursuing domestic AI development to mitigate geopolitical risks. While their current quality may lag behind global leaders, the state-level strategic initiative is driving rapid capacity expansion.
• Governance Reform: Japan is highlighted as a uniquely attractive market due to the confluence of AI-related hardware dominance and ongoing corporate governance reforms that prioritize shareholder value.

5. Notable Quotes

• Ning Hui Liu (State Street): "The real hidden reality is AI is not just a software story, but it's very much a physical, a very Asian-hardware story as well."
• James Smith (Palliser Capital): "You can identify these choke points in the region specifically where pricing increases are possible, that very large expansions, very large order sets are coming through."

Synthesis

The AI boom has transitioned from a software-led narrative to a hardware-intensive reality, positioning Asia as the indispensable engine of the global AI economy. The most significant investment opportunities are no longer just in the "hyperscalers," but in the upstream material and component suppliers—many of which are legacy industrial firms that have pivoted to provide critical, high-margin components like ABF substrates and advanced ceramics. As the supply chain tightens, the ability to scale manufacturing capacity and navigate geopolitical bifurcation will determine the long-term winners in the region.