The software sector currently stands at the convergence of two major transformative waves: the "AI industrial revolution" and the "policy dividends from domestic technology innovation." The core investment drivers can be broken down into four key dimensions.
Firstly, AI is accelerating growth, while cloud computing price increases support profitability. The year 2026 is seen as the starting year for AI Agent commercialization, with token consumption growing exponentially. Each token consumed directly monetizes software capabilities. The new "pay-per-process/result" business model of AI Agents, compared to traditional annual or per-seat software licenses, not only creates new markets but also reallocates existing budgets for traditional outsourcing and software procurement, potentially opening a market far larger than traditional software. Concurrently, cloud service providers globally are entering a price hike cycle, and the value reassessment of AI computing power and cloud services further strengthens the pricing anchor for software services.
Secondly, domestic innovation policies are driving accelerated order releases. Document No. 79 from the State-owned Assets Supervision and Administration Commission (SASAC) mandates that by the end of 2027, all central and local state-owned enterprises must achieve 100% replacement of their information systems with domestic innovative technologies. Government procurement favoring "domestic products," through lists, price incentives, and standardized assessments, is guiding concentrated demand release, pushing quality domestic tech products from "usable" to "large-scale application." With the 2027 deadline approaching, 2026 is becoming a critical window for the concentrated landing of domestic tech orders.
Thirdly, breakthroughs in core technologies are opening up industry elasticity. Policy focuses on overcoming bottlenecks in areas like basic software with extraordinary support, promoting deep integration of "cloud-native, AI-native, and domestic tech-native" approaches. This breaks the fragmented pattern of point solutions and builds a self-reliant, collaboratively evolving industrial ecosystem. Meanwhile, increased participation from private and small-to-medium enterprises is boosting industry vitality and flexibility.
Fourthly, the sector is entering a valuation repair window with a margin of safety. Previous narratives about "large models devouring software" led to significant valuation discounts for the software sector, but this involved two major misjudgments. First, surging spending on computing power has not squeezed corporate IT budgets; instead, AI Agent applications are creating new incremental budgets (e.g., in marketing). Second, while AI coding reduces code costs, the value of non-code elements like data moats increases, strengthening the competitive advantages of application providers. In the AI era, the pricing logic revolves around tokens, creating a revaluation window for software sectors with high growth potential and currently low valuations.
Introducing the Huabao Software Development ETF
The Huabao Software Development ETF (Ticker: 159036) passively tracks the CSI All Share Software Development Index (Ticker: 932094.CSI). It provides 100% exposure to the software development industry with 112 constituent stocks, aiming to reflect the overall performance of software development companies within the CSI All Share Index universe. While the software sector is in an upward cycle with immense potential, competition is fierce. In an environment where it's unclear which specific sub-sector or individual stock will outperform, a broader, more diversified approach can capture the beta returns from the overall software industry's development.
Key Advantages of the Underlying Index
The underlying CSI Software Development Index is characterized by its pure sector focus and high diversification. Compared to similar software indices, it has three clear advantages: 1) Purer industry focus—100% concentrated on software development. 2) More comprehensive holdings—includes 112 constituent stocks, dispersing single-stock risk. 3) More balanced allocation—the top ten holdings account for about 40% of the weight, with a single-stock cap of 10%. During the software industry's upcycle, this broader, more balanced layout aims to capture industry beta returns driven by multiple forces like AI, domestic tech innovation, and industrial software.
Specifically, in terms of industry distribution, the index is 100% allocated to software development. In contrast, other software indices have lower weights in this sector. Regarding constituent count, this index has 112 stocks, while comparable indices typically have 50 or 30. The combined weight of its top ten holdings is approximately 42%, compared to over 50% or even 60% for some peers. Its single-stock weight cap is 10%, lower than the 15% cap common in other software indices.
Highlights of the Software Development Index
The index encompasses several hot themes. The weight of stocks associated with AI applications, cloud computing, the domestic tech innovation industry, fintech, cybersecurity, and the HarmonyOS ecosystem is significant.
In terms of performance, since 2023, the index's annualized return has outperformed several comparable software indices.
Its top ten holdings include leading companies in the software space such as iFLYTEK Co., Ltd., Zhejiang Hithink RoyalFlush Information Network Co., Ltd., Kingsoft Office Corporation, Talkweb Information System Co., Ltd., Hundsun Technologies Inc., 360 Security Technology Inc., Compass Inc., HopeRun Software Co., Ltd., Sangfor Technologies Inc., and Thundersoft Technology Co., Ltd., with a combined weight of 42.07%.
Reassessing the "Large Models Devouring Software" Narrative
The SaaS (Software-as-a-Service) industry traditionally operated on subscription models. In contrast, AI-native companies often use usage-based or outcome-based pricing, such as charging per million tokens or for task completion. This shift, accelerated by the commercial success of companies like Anthropic, led to peak concerns about "large models devouring software."
However, software companies are not passive; they are building core moats through transformation. These include data moats from accumulated industry and user data, complex process moats in customized workflows, and industry know-how moats in vertical sectors like energy and industry. The value lies in deep domain understanding and application, requiring integration with regulations and business practices.
From an earnings perspective, the dense release of Q1 2026 reports has gradually alleviated these concerns. Several domestic software firms with vertical moats have delivered better-than-expected results driven by AI-powered product upgrades and accelerated commercialization, marking a key shift from "AI disruption" to "AI empowerment." For instance, Kingsoft Office reported AI subscription revenue surging over 300% year-over-year, with AI feature monthly active users exceeding 120 million and paid conversion rates rising steadily. The company stated that AI is not only a core driver of user growth but also a key variable in increasing average revenue per user.
Essentially, this is not a story of disappearance but of upgrade. The software sector faces a business model upgrade, not the demise of traditional software. This includes pricing model upgrades (from per-seat/annual to usage/outcome-based), value proposition upgrades (from function-driven to value-driven), the opening of new incremental spaces (AI Agents creating new markets and reallocating existing budgets), and a strengthened pricing anchor (from cloud provider price hikes and the value reassessment of AI compute).
In summary, AI is not devouring software. Instead, it is comprehensively reconstructing the software industry's business models and value distribution rules, with tokens as the pricing unit and Agents as the new form. Software companies with data moats, process depth, and industry know-how see their competitive advantages not eroded but potentially amplified in the AI era.
Current Valuation of the Software Development Index
Earlier this year, the AI application sector saw significant valuation discounts compared to computing power and model layers due to the "large models devouring software" narrative. Some analysts believe this pricing logic is a misjudgment. In the AI era, pricing revolves around tokens, creating a three-tier markup structure (compute, model, application). The value reconstruction of scenario-specific tokens at the application layer may not be fully priced in by the market. End enterprise users buy business outcomes, not tokens. The AI application layer, empowered by private data and industry know-how, allows generic model tokens to gain greater value, enabling higher token prices. High-growth, low-valuation AI application targets with existing token-consumption or outcome-based revenue streams may attract market attention.
Since its high on January 14, 2026, the Software Development Index has declined significantly, currently underperforming AI hardware sectors. This positions the software板块 as a segment with relatively lower "water level" and higher risk-reward potential within the AI产业链—having limited downside but strong upside elasticity if sentiment turns. As of early June, the index's P/E ratio is at a historically low percentile, indicating relatively high valuation attractiveness and a margin of safety.
The Software Spring Following Hardware
Reviewing past technological revolutions, profits have consistently flowed from hardware to applications. Hardware opens the track, but software more sustainably harvests profits. The second half of the AI revolution likely lies in applications. Historical analysis of tech cycles like the internet, mobile internet, and cloud computing shows they are often triggered by hardware breakthroughs, laying the foundation for subsequent software transformations. Software vendors' fundamental improvements typically lag behind hardware suppliers, but their revenue growth tends to be more sustainable.
Reflecting on the mobile internet path, infrastructure is built first, followed by device普及, and then a百花齐放of applications. The current scenario, with computing power leading and applications lagging,基本符合this objective industry cycle规律; hardware outperforming first is a normal phenomenon.
Computing power and AI applications are two essential stages of a complete industry cycle.回顾past major bull markets in the TMT sector, by the end of those cycles, the performance gaps between sectors like electronics, communications, computer, and media were usually not extreme. Currently, hardware-related sectors have seen larger gains, while AI application-related sectors like computer and media lag. Historical experience suggests these performance gaps may "converge." The software development sector related to AI applications still holds high allocation value. Especially in the mid-to-later stages of a行情, as incremental capital enters, earlier high-flying sectors may face "profit-taking," while the lower-positioned software development sector, supported by the industrial logic of AI implementation and attractive valuations, coupled with prior underperformance, may更容易attract capital inflows.
How Domestic Tech Innovation Policies Benefit the Software Sector
Policy mandates are driving order releases, with 2026 being a crucial year ahead of the 2027 full replacement deadline. SASAC's requirement for 100% domestic tech replacement in SOEs by end-2027 is pushing渗透from government and party offices to eight key industries like finance, telecom, and energy, and from peripheral to core business systems. Detailed government procurement standards further solidify the replacement pace.
Funding is being secured through the expansion of ultra-long-term special government bonds and local debt resolution plans, providing a financial保障for domestic tech procurement. The bond proceeds target major national strategic, security, and new领域, with domestic tech innovation as a core area for科技安全, data security, and supply chain security. Local debt resolution frees up budgetary space.
Market size is substantial. Predictions suggest the domestic tech innovation industry scale in China could exceed significant figures by 2026, with high compound annual growth rates, and approach a multi-trillion规模by 2029.
Performance is being validated as policy expectations translate into real corporate orders and earnings. For example, some companies have reported growth in contracts from central and state-owned enterprise clients.
Analyst views highlight that over 70% of the software sector's growth may be contributed by domestic tech innovation, with 2026 being a peak year for earnings realization. Domestic tech innovation is seen as the core growth engine for the sector, with AI providing弹性补充. It is considered the core beta driving sector景气度and valuation, while AI is the core alpha driving弹性and growth space.
Investment Opportunities in AI-Reshaped Software Sub-Sectors
Under the trend of AI reshaping software, the value of vertical "AI+Software" applications is becoming prominent. Compared to the randomness and non-standardization of general large model responses, AI applications in verticals like finance, healthcare, and e-commerce, through预设scenarios, optimized prompts, and专属data training, can provide more accurate, stable, and practical solutions. The deep integration of "AI+Software" is creating real value across industries.
In AI+Finance, tools like market software, investment research tools, and robo-advisors are rapidly integrating AI capabilities. In an A股environment盛行with thematic investing and new concepts, financial AI tools that can quickly interpret information and梳理industry chains have clear application scenarios and growth potential.
AI+Healthcare covers areas like AI-assisted triage, imaging辅助diagnosis, and drug molecule R&D. AI can significantly shorten R&D cycles and improve diagnostic efficiency, representing a highly certain融合direction. While mature business models are still being explored, continuous attempts and technological progress pave the way for long-term development.
AI+Other Verticals: Similar transformations are occurring in e-commerce, education, government affairs, industrial software, and other fields. The process of software companies achieving product upgrades and value reassessment by introducing AI capabilities has just begun.
Analyst research points out that software companies with industry know-how and high-quality data are likely to build moats and fully benefit from the value growth brought by AI empowerment in the AI Agent era. Some brokerages list AI applications and domestic computing power as dual main investment themes for 2026, focusing on sub-sectors like AI+Finance, AI+Industrial Software, AI+Healthcare, AI+Office, AI+ERP, and AI+Marketing.
Product Details and Fees
The Huabao Software Development ETF (159036) is set to list on June 11th. For specific product details, fees, and comprehensive risk information, investors should refer to the official fund documents such as the Fund Contract, Prospectus, and Key Facts Statement. The ETF tracks the CSI All Share Software Development Index. Management is by Huabao Fund. Investors are advised to carefully read the fund legal documents, understand the fund's risk-return characteristics, and choose products suitable for their own risk tolerance. Fund investments carry risks.
Comments