If the first half of the AI race was defined by computational power, what defines the second? On August 31, 2026, a monkey captured primetime television audiences. This wasn't a rerun of "Journey to the West," but the domestic debut of "Journey to the West: The Later Chapters," an AIGC long-form series airing simultaneously on Hunan Satellite TV and Mango TV. Its first season of 30 episodes was largely generated by AI, from scriptwriting to visuals. While this "monkey" is just a show, it represents a clear signal: AI applications are moving from the laboratory to the living room, transitioning from corporate presentations to tangible revenue on financial statements.

The first half of the year was spent "buying shovels," but now the gold mine's outline is emerging. Can AI applications carry the momentum into the second half? Four factors are converging: the rotation of capital away from crowded hardware trades, software companies converting their AI revenue from promises to profit, the resilience of asset-light software compared to capital-intensive chips in a high-interest-rate environment, and the shift of profits toward applications as large models become more powerful and affordable.

Overseas: Shovels Rest, Gold Rush Begins

Overseas markets are showing a dramatic divergence. While capital was recently pouring into Nvidia's chips, the Philadelphia Semiconductor Index has pulled back since August, even as North American software stocks surge. Capital is flowing from the "shovel-selling" hardware sector to software applications that are actually "finding gold." The reason? Overseas software giants have finally turned AI from a story into a profitable one.

For example, Palantir, once considered a "government project company," saw revenue surge 89% year-over-year in the first half of 2026, with its AIP-related enterprise agents moving from proof-of-concept to formal procurement. Meanwhile, Salesforce's Agentforce has surpassed $1.5 billion in annualized recurring revenue, up roughly 240% year-over-year, with over half of new orders coming from existing customers adding more services.

This reveals the core logic behind the overseas AI application boom: AI isn't disrupting SaaS; it's increasing its value. These software providers, with a decade or two of enterprise service experience, hold the most valuable assets—proprietary data, customer relationships, and business processes. General-purpose large models can only achieve a 60-point solution, but these companies deliver over 95 points for their clients. Their business models are evolving from simple "per-seat subscriptions" to a hybrid pricing structure of "seats + AI consumption + value-added modules," effectively building a new revenue layer on top of existing subscriptions. AI hasn't killed SaaS; it has become its most valuable enhancement layer.

This points a clear direction for the global AI industry: rather than competing on foundational models, focus on who possesses irreplaceable industry know-how.

Domestic: From Following to Leading

This encouraging overseas news is resonating back home. During the 2026 interim earnings season, pure AI application companies delivered impressive results, proving that "AI is beginning to make money" is no longer a slogan but a fact on the balance sheet.

In the enterprise AI market, the star company MeiFuShi saw its first-half revenue more than double year-over-year, with net profit soaring over fourfold, while its AI-related business grew 123.7%. Established software vendor Kingdee International reported a 189% year-over-year increase in its AI-native product revenue, including a 283% jump in subscription revenue. Kingsoft Office also observed a significant wave of additional purchases from existing enterprise clients for its organizational-level AI products. The underlying logic is consistent: domestic companies are now willing to pay for AI agents that genuinely reduce costs and improve efficiency, and AI's penetration in the B2B sector has entered a phase of performance delivery.

In AI content production, Chinese companies are even leading the world. On August 31, the domestic AIGC series "Journey to the West: The Later Chapters" aired during primetime, a symbolic milestone as AI content moves from small-screen experiments to mainstream large-screen media. Furthermore, domestic animated dramas grew over 100% year-over-year in the first half, Kunlun Wanwei's AI short-drama platform saw revenue rise about 160%, and China Online also achieved high growth in its short-drama business, with its overseas platform already profitable.

Why did AI content take off in China first? Because the country boasts the most complete short-video industry chain and the most mature user acceptance. Since August 31, related companies like Mango Excellent Media and China Online have continued to climb, which is the market's vote for this logic.

Related ETFs for the Second Half

The investment logic for the second half is clear: hardware provides foundational support, while applications contribute incremental growth. For average investors, betting on a broad basket via ETFs might be a better way to diversify risk and capture trends than picking individual stocks.

The ChiNext Artificial Intelligence ETF (159363) offers a dual-driver approach with optical modules and AI applications. Its underlying index includes Zhongji Innolight, Eoptolink, and Tianfu Communication at over 35% combined, focusing on optical module and CPO leaders as core AI computing power players. The feeder funds are A Class 023407 and C Class 023408.

The Fintech ETF (159851) combines internet finance and AI applications, as finance is poised to be one of the fastest areas for AI implementation. Its index is heavily weighted in computers and non-bank financials, covering internet brokers, fintech IT, cross-border payments, and AI applications, offering a mix of financial cycle and tech growth attributes. The feeder funds are A Class 013477 and C Class 013478.

The Technology ETF (515000) serves as the "Pro Max" version of a broad tech fund, a core tool for tech sector investing. It gathers 50 A-share tech leaders, combining "hard tech beta" with "quality leader alpha," and includes leaders in optical modules, semiconductor equipment, memory chips, and computers. The feeder funds are A Class 007873 and C Class 007874.

Data source: Shenzhen Stock Exchange, Shanghai Stock Exchange, Wind, etc.

Reminder: Market fluctuations may be significant, and short-term performance does not predict future returns. Investors should make rational decisions based on their own financial situation and risk tolerance, paying close attention to position and risk management. According to the fund manager's assessment, the Fintech ETF and Technology ETF are rated R3-Medium Risk, suitable for balanced (C3) and above investors. The ChiNext AI ETF is rated R4-Medium-High Risk, suitable for aggressive (C4) and above investors. For an appropriate match, please consult your sales institution.

Risk Disclaimer: The ChiNext Artificial Intelligence ETF passively tracks the ChiNext Artificial Intelligence Index, with a base date of December 28, 2018, and a release date of July 11, 2024. The Fintech ETF tracks the CSI Fintech Theme Index, with a base date of June 30, 2014, and a release date of June 22, 2017. The Technology ETF tracks the CSI Technology Leader Index, with a base date of June 29, 2012, and a release date of March 20, 2019. Index constituent stocks are adjusted according to the index's compilation rules, and backtested historical performance does not guarantee future results. The index constituents mentioned are for demonstration only, and descriptions of individual stocks are not an investment recommendation nor do they represent the holdings or trading activities of any fund under the management company. Any information in this article (including but not limited to stocks, comments, forecasts, charts, indicators, theories, and any form of expression) is for reference only. Investors are solely responsible for their own investment decisions. Furthermore, any views, analyses, or forecasts in this article do not constitute investment advice to readers and are not liable for any direct or indirect losses resulting from the use of this content. Fund investment carries risks; past performance does not signify future results. The performance of other funds managed by the fund manager does not guarantee the performance of this fund. Please invest cautiously.