China IT Monthly: Dual-Engine Era of Hardware Scale and AI Commercialization

Bottom Line

This month, the key trajectory of China's IT industry has evolved from singular technological breakthroughs to a clear resonance and mutual validation between two major growth engines: "Domestic Hardware Scaling" and "AI Application Commercialization." Investors should adjust their positions accordingly: immediately overweight the "picks-and-shovels" beneficiaries (equipment, materials, advanced manufacturing) directly profiting from the semiconductor industry-wide price hikes and capacity expansion, while simultaneously increasing holdings in AI application-layer platforms that have demonstrated large-scale user acquisition and monetization capabilities. Avoid pure AI concept stocks reliant solely on market sentiment without substantive revenue support.

Month in Review

February 2026 presented a clear and multi-layered narrative arc for China's information technology sector. Early in the month, the narrative core focused on the scaled breakthrough of hardware self-sufficiency: the Nanjing software cluster's revenue exceeding ten trillion yuan marked a new magnitude for the industry's agglomeration effect. Meanwhile, the aggressive 80% semiconductor self-sufficiency target, massive capital expenditures by fabs, and industry-wide price hikes collectively formed a solid foundation for the revaluation of the hardware supply chain. Market sentiment continued to warm up amid clear policy support and strong financial signals.

From mid-month to month-end, the narrative underwent a critical evolution and intensification. On one hand, strong signals from the semiconductor sector extended from capacity and prices to capital markets (massive financing, soaring IPO valuations) and cutting-edge technology (quantum communication chip networks), indicating a comprehensive acceleration from quantitative to qualitative change. On the other hand, the more pivotal turning point was in AI: the explosive user growth of leading AI applications during the Spring Festival marked the first time large language model technology achieved "societal-level penetration," formally transitioning from labs and B2B testing to large-scale B2C and B2B commercialization and monetization. Concurrently, the deep adaptation between "domestic AI chips and domestic large models" filled a critical gap in the closed-loop ecosystem for autonomous, controllable computing power.

The story of the entire month was a transition from "determined investment" to "fruition." The semiconductor industry, nurtured by years of policy and capital, began collectively entering a harvest period (profit recovery, capacity ramp-up, technological breakthroughs). Meanwhile, the AI industry crossed the threshold of technological spectacle, entering a new phase of user value and commercial value validation. Together, they defined the dual-engine driving paradigm for China's IT growth this month—and for the coming cycle.

Trajectory Analysis

Month-over-Month Change: Significantly Better. Compared to the previous month (inferred from weekly reports), this month saw positive evolution across multiple dimensions: 1) Industry prosperity shifted more from expectations (policy, targets) to financial realities (industry-wide price hikes, record revenues from leading companies); 2) Technological breakthroughs expanded from single links (e.g., chip design) to system-level achievements (quantum communication networks, hardware-software ecosystem adaptation); 3) Investment themes differentiated from generalized "independent innovation" into two solid, trackable pillars—the scaled realization of semiconductor localization and the commercialization launch of AI applications. The visibility and certainty of the overall trajectory improved substantially.

Key Developments

1. AI Applications Achieve "Nationwide Penetration," Commercialization Inflection Point Established (Week 3) — During the Spring Festival, the daily active user count for leading tech companies' AI applications surged dramatically, e.g., Tencent's "Yuanbao" DAU exceeded 50 million. This marks the first time China's large model technology has crossed the chasm from tech demos to societal-level applications, entering the critical stage of user value creation and commercial monetization validation. Portfolio implication: Investment focus should tilt towards the application layer. Prioritize increasing holdings in internet platforms and vertical SaaS leaders with massive existing user bases that can efficiently convert AI traffic into value-added service revenue. Simultaneously, decisively reduce or avoid pure AI technology companies still in high R&D investment phases, with ambiguous business models and unclear monetization paths. Value capture at the application layer will precede and potentially outperform the foundational model layer.

2. 80% Semiconductor Self-Sufficiency Target Resonates with Industry-Wide Price Hikes (Week 1 & 2) — Domestic semiconductor self-sufficiency is projected to jump to 80% in 2026. Concurrently, since the start of the year, widespread price increases have occurred across the entire industry chain from memory chips to analog/power chips, packaging & testing, with some products seeing hikes up to 80%. This represents both an aggressive policy target and a direct financial signal of a reversal in market supply-demand dynamics. Portfolio implication: This constitutes a "policy + fundamentals" double-hit logic for the semiconductor sector. Immediately overweight segment leaders with pricing power and capacity advantages during the price hike cycle, especially analog chip, power semiconductor, and memory chip design companies. Furthermore, the core support for the self-sufficiency target lies in manufacturing capability; thus, semiconductor equipment, materials, and foundry leaders will continue to receive confirmed orders, making them core long-term allocation directions.

3. Nanjing Software Industry Scale First Exceeds Ten Trillion Yuan; Data Element Policy Framework Set (Week 1) — Nanjing became the first city cluster with software business revenue exceeding ten trillion yuan. Simultaneously, the National Data Bureau and other departments issued documents providing an institutional framework for data circulation and trading. This provides dual endorsement for the sustainable development of the software and data services industry from both industrial scale and institutional foundation dimensions. Portfolio implication: Validates the超额 growth potential of software industry agglomeration. Focus on allocating to software and solution providers with headquarters or core R&D in Nanjing and the Yangtze River Delta, possessing deep industry know-how. Concurrently, data element policies directly benefit data governance, privacy computing, data security, and data trading platform operators. Screen for companies already participating in key industry or government data pilot projects.

4. "Domestic AI Chip + Domestic Large Model" Deep Adaptation Completed (Week 3) — Domestic AI chipmaker Taichu Electronics announced its chips have completed deep adaptation for mainstream large models like Zhipu AI's GLM-5.0 and Alibaba's Qwen3.5. This marks a key advancement in the ecological synergy between autonomous computing power and autonomous algorithms, laying the foundation for building an AI tech stack free from external constraints. Portfolio implication: Investment logic upgrades from "point replacement" to "ecological synergy." Focus on挖掘 and positioning AI chip design companies, system integrators, and middleware providers with unique technological barriers in hardware-software co-optimization. Such enterprises will hold significant first-mover advantages and准入壁垒 in future government, state-owned enterprise, and critical industry procurement.

5. Integrated Circuit Industry Financing Exceeds 80 Billion Yuan; Montage Tech IPO Sees Significant Valuation Premium (Week 2) — Over the past 14 months, total industry financing reached 83.5 billion yuan, with AI chip companies particularly active. Meanwhile, semiconductor design company Montage Technology's Hong Kong IPO saw its market cap突破 210 billion yuan on the first day, indicating ample primary market funding and extremely high valuation premiums given by the secondary market to core technology companies. Portfolio implication: Massive continuous capital inflow will accelerate technology iteration and优胜劣汰. Investors can broadly position in innovation clusters in frontier fields like AI chips through thematic ETFs or primary market funds-of-funds. Montage Tech's case provides a positive valuation anchor for即将上市 or already-listed technology leaders in细分赛道. Focus on companies with similar "hidden champion" potential in emerging high-growth areas like high-speed interconnect and automotive chips.

6. Semiconductor Manufacturing Leader Hits Record Revenue, Continues Capacity Expansion; Industry Consolidation Accelerates (Week 3) — SMIC reported record revenue for 2025 and continues to expand 12-inch wafer capacity. Concurrently, Hua Hong plans to acquire HLMC, and AMEC intends to acquire Zhonggui Electronics, showing industry leaders expanding through both organic growth and M&A. Portfolio implication: Healthy financials and expansion plans from leading companies provide clear order visibility for the entire upstream supply chain (equipment, materials, components). Simultaneously, the industry consolidation trend is clear. Focus on medium-sized "specialized, sophisticated, distinctive, and innovative" enterprises that could become M&A targets or industry consolidators, especially those with unique technologies in细分领域 like semiconductor equipment and special process materials.

Risk Evolution

Risks That Materialized: Some "Valuation Bubble Risk" materialized this month, evident in the frenzied rise of个别 AI star stocks detached from fundamentals, contrasting with the月初 market optimism about improving industry fundamentals. New Emerging Risks: Risk of Intensifying Divergence in Intra-Industry Prosperity. This month's information clearly shows prosperity shifting from "broad-based gains" to "structural boom." Resources (capital, talent, orders) are高度集中 towards leading enterprises that have achieved technological breakthroughs, possess pricing power, or have validated business models. Mid- and tail-end companies without core competitive advantages may face risks of market share and profit margin compression.

Sector Pulse (Monthly)

Outlook: Next Month

Key catalysts to watch:

• Follow-up details on key financial reports and M&A cases from SMIC, Hua Hong Semiconductor, etc.: The market will scrutinize their capital expenditure efficiency, gross margin changes, and M&A integration plans to judge the盈利质量 and future growth guidance of the wafer manufacturing segment.
• Sustainability and scope of the semiconductor price hike潮: Monitor whether price increases further transmit from the design and manufacturing links to broader component areas, and the acceptance level of downstream end customers. This will verify the strength and sustainability of the industry recovery.
• Preliminary data on AI application monetization: Following the Spring Festival traffic peak, key operational metrics for mainstream AI applications—user retention rates,付费转化率, and ARPU (Average Revenue Per User)—will face market scrutiny for the first time, determining whether the valuation logic for application-layer companies continues upward or faces a correction.

Positioning recommendation: Based on the dual themes of "Hardware Scaling + Software Commercialization," implement a barbell strategy. Overweight semiconductor equipment, materials, and manufacturing/packaging leaders with capacity advantages on one end. Increase holdings in AI application platforms and industry software solution providers with clear business models and massive user bases on the other end. Strictly control exposure to纯 AI concept stocks with extremely high valuations; consider using their volatility for tactical hedging. The overall portfolio should tilt towards the certainty of the mid-stream (manufacturing, applications) while balancing the long-term growth of the upstream (equipment, materials).