Explosive! The STAR Artificial Intelligence and STAR Chip sectors have surged in a reversal, the "dark side of light" has erupted, and Hwabao Fund's Electric Utilities ETF closed at a new high! Hong Kong's internet sector is brewing a turnaround

On May 19th, the A-shares staged a dramatic reversal, with the three major indices opening lower and fluctuating in the morning, then gaining momentum in the afternoon to recover losses. By the close, the Shanghai Composite Index rose 0.92% to 4169.54 points, while the ChiNext Index slightly fell 0.16% to 3908.44 points. The total trading volume of the two markets reached 2.89 trillion yuan, with over 3,600 stocks rising and more than a hundred hitting the daily limit. The chip and semiconductor sectors surged, with the Hwabao CSI STAR Artificial Intelligence ETF (589520) focusing on domestic AI chains seeing its price soar 4.21% close to previous highs, while the Hwabao CSI STAR Chip ETF (589190), which is fully invested in the chip industry, rebounded to rise 3.82%, with a daily surge of 7.48% stabilizing the upward trend. The “light” and “electricity” sectors showed short-term differentiation, with the power sector experiencing a wave of limit-ups. The Hwabao Power ETF (159146), which covers the entire energy layout of “wind, solar, water, fire, and nuclear,” closed with a gain of 3.63%, setting a new historical closing high! The CPO of optical modules weakened, and the Hwabao ChiNext Artificial Intelligence ETF (159363), which contains over 50% optical module CPO, saw its price drop over 3% at one point, closing down 1%, with funds increasing positions by 140 million shares during the dip. Brokerages showed unusual movements in the afternoon, with the Hwabao Top-tier Brokerage ETF (512000), valued at 36.3 billion, suddenly rising to close up 1%. On May 18th, CICC announced the latest progress on the merger with Dongxing Securities and Cinda Securities, with the merger expected to make it the fourth brokerage in the industry to enter the “trillion club.” Hong Kong stocks released warmth, with internet leaders rebounding. Tencent Holdings rose over 4% during the session, closing up more than 2%, while the Hwabao Hong Kong Internet ETF (513770), a core tool for AI in Hong Kong stocks, increased nearly 1% from the bottom. Recently, the AI market has continued to expand, with several institutions suggesting that low-priced Hong Kong internet stocks are likely to seize opportunities from the AI expansion. The consumption of tokens continues to rise, providing a definite benefit to cloud vendors like internet leaders. Alibaba and Tencent are currently in a critical breakthrough period for transforming into “AI + Cloud” technology platforms, likely to welcome value reassessment. 【ETF All You Need to Know Hotspot Review】Next, let's focus on the trading and fundamentals of the sectors of AI innovation, electricity, and Hong Kong internet.】 【Domestic AI strong players remain strong! Chipone Technology tops the A-share fundraising list, and Hwabao Fund's CSI STAR Artificial Intelligence ETF (589520) surged 4.21% to challenge previous highs!】 Continue the music and dance! The direction of self-control surged in the afternoon, with Chipone Technology leading the rise by over 15%, hitting a historical high, Amlogic rose by more than 12%, and Cambricon Technologies increased by over 8%. The Hwabao SSE STAR Artificial Intelligence ETF (589520), which focuses on the domestic AI industry chain, saw its intra-day increase peak at 4.34%, closing up 4.21%, just a step away from the listing high point (0.774) on May 14, with a total transaction volume of 62.46 million yuan, a 14% increase compared to the previous period. Why is the direction of AI in the science and technology innovation sector continuously surging, like a bamboo shoot after rain? This can be analyzed from three levels: 1. Policy level, high-level guidance on computing power network construction. Senior leaders pointed out that computing power is an important embodiment of national comprehensive strength, and the construction of a national integrated computing power network should be accelerated. Simply put, the computing power network is like a “computing power version of the national grid”, where one can “buy computing power” on demand, paying as needed, as conveniently as using water or electricity. Industry insiders noted that the construction of the computing power network, like building high-speed rail or 5G networks, has a strong “multiplier effect” that can drive the upgrade of the entire industrial chain, including artificial intelligence, big data, and industrial internet. 2. Industry level, the Google Developer Conference (Google I/O Conference) is being held today, and AI on the edge is expected to receive attention. Zhongyin Securities stated that leading companies' continuous investment in model lightweighting and edge computing power is gradually lowering the threshold for terminal deployment, and providing clearer technical paths and cooperation foundations for chips, modules, terminals, and applications, benefiting AI on the edge continuously. 3. Individual stock level, the heavyweight stock Chipone Technology achieved a new high in new orders signed in Q1 2026, fully validating the ASIC industry trend, benefiting from the construction of cloud AI infrastructure and the popularization of edge AI hardware applications. In Q1 2026, companies like Amlogic and other edge SoC manufacturers performed well, driven by the dual forces of new product volume and effective price transmission, with many manufacturers achieving high revenue growth, further confirming the upward trend of edge AI hardware prosperity, resonating with Chipone's edge business. Looking ahead, Dongxing Securities believes that the mid-term technology mainline configuration logic has not changed. Unless there are significant reductions in capital expenditures by large companies or financing difficulties, the potential orders for 2027 and opportunities for upward revisions of performance expectations are still expected to drive the market further, with core technology varieties remaining a strong direction in terms of overall market prosperity in the mid-term * [The Light of Domestic Substitution, Self-reliance and Strength in Science and Technology] The Hwabao (589520) Science and Technology Artificial Intelligence ETF and its linked funds (Linked A: 024560, Linked C: 024561) focus on the domestic AI industry chain, with constituent stocks including domestic GPU leaders (such as Cambricon), domestic ASIC leaders (such as Chipone), AI application leaders (such as Kingsoft Office), with nearly half of the weight in the semiconductor industry, demonstrating strong aggressiveness; the software industry accounts for over 30%, expected to benefit from the AI application rebound. Meanwhile, this ETF is a margin trading target, serving as an efficient tool for one-click allocation of domestic computing power. ["Light" Adjustments, "Electricity" Defense! Hwabao Fund Electric Power ETF (159146) Rises 3.63% to Reach New High, 8 Stocks Hit Limit Up! Collaborative Synergy of Computing and Electricity Continues to Land] The electric power sector has exploded with limit-up trends, with the CSI All Share Electric Utilities Index rising over 3.5%. All constituent stocks surged, with 8 stocks hitting the limit up, including Fuling Electric Power, Shanghai Electric Power, Jiantou Energy, Ganeng Co., Ltd., YN Energy Holdings, Guangxi Energy, Linyang Energy, and Huaneng Mengdian, while multiple stocks such as Zhejiang Energy Power, Beijing Energy Power, Jiangsu New Energy, Datang Power, and Guodian Power surged over 5%. In terms of popular ETFs, the Hwabao Electric Power ETF (159146), which has a full energy layout of "wind, solar, water, fire, and nuclear," strengthened throughout the day, closing up 3.63% and reaching a new high since listing! Daily trading volume exceeded 44 million yuan, with funds accumulating over 42 million yuan in the past five days. In contrast, the "light" sector weakened, with the Hwabao (159363) ChiNext Artificial Intelligence ETF, which has over 50% CPO content, falling 1% in the market, while funds increased their holdings by 140 million shares during the dip. In terms of market structure, A-shares often switch between "light" and "electricity": today, the optical module sector adjusted along with external factors, while electricity remained strong independently, reflecting electricity's defensive attributes within computing power and revealing the core logic that distinguishes electricity from traditional dividend sectors—in the AI era, electricity also possesses certain growth potential. From the news perspective, today the power sector experienced a surge in limit-up stocks, driven primarily by the "computing power and electricity synergy" moving from policy expectations to actual implementation. The direct news catalyst may come from two aspects: first, the domestic realization of large data centers participating in electricity spot trading in the form of virtual power plants, achieving "computing power following electricity", which validates the business model of bidirectional synergy between computing power and electricity; second, the Shanghai press conference clearly stated that the "14th Five-Year Plan" aims to double computing power scale and build four major green electricity bases of tens of millions of kilowatts, strengthening demand expectations. Galaxy Securities stated that power companies are laying out computing power and electricity synergy, and the industrial trend is gradually being established. It is recommended to grasp two main lines of power investment: 1. Entering the data center track to create a second growth curve for companies; 2. Supplying power to data centers, with existing projects already implemented, or companies that have layouts in core areas such as the eight major computing power hub nodes with positioning advantages, are expected to benefit from dual improvements in volume and price in the future. In the AI era, the role of "selling shovels" has initiated the revaluation of the power sector's value. Power ETF Hwabao (159146) and its connected fund (code: 026949) focus on the public utility sector, comprehensively laying out thermal power, hydropower, wind power, nuclear power, and photovoltaics, capturing the growth elasticity of new energy installations while relying on the high dividends and stable cash flow of traditional power leaders to smooth market fluctuations, achieving a dual adaptation of "defensive base + growth elasticity." 【AI market diffusion, Hong Kong stocks internet brewing a turning point! Hwabao Fund 513770 rises nearly 1% from the bottom, focusing on value revaluation under the Token craze】 Hong Kong stocks' AI rebounded from low levels, with most internet leaders rising, Tencent Holdings rose over 4% during the session, closing up more than 2%; Alibaba-W and Meituan-W rose over 1%. The core tool of Hong Kong stocks AI—Hong Kong stocks internet ETF Hwabao (513770) opened low and rose high, maintaining an upward trend throughout the day, with an intra-day price increase of 0.99%. Just the day before, Hong Kong stocks internet ETF Hwabao (513770) dipped to a new low in this round of correction, with its tracked CSI Hong Kong Connect Internet Index currently having a price-to-earnings ratio PE (TTM) of only 20.06 times, located in the historical bottom range at the 4.3% percentile over the past 10 years, providing a high safety margin. Recently, as the AI market spreads, multiple institutions have indicated that after a significant rebound on the hardware side, the Hong Kong internet sector is expected to welcome market opportunities brought by the diffusion of the AI industrial chain. Industrial Securities pointed out that the Hong Kong internet sector, due to its trading attributes of domestic computing power and the direction of AI applications benefiting first, is expected to become one of the important main lines of market diffusion In addition, from the perspective of overseas mapping, cloud vendors represented by Hong Kong stock internet are also expected to be the direction for subsequent domestic AI market rebound. According to the latest calculations by OpenRouter, the total call volume of global AI large models reached 26.9 trillion tokens last week (from May 11 to May 17), marking a continuous four-week increase. Among them, the weekly call volume of Chinese AI large models reached 7.693 trillion tokens, which is 1.81 times that of the United States during the same period, surpassing it for three consecutive weeks and maintaining the top position globally. The token consumption continues to rise, which constitutes a definite benefit for cloud vendors such as Alibaba and Tencent. At the same time, the commercialization of AI by leading internet companies is accelerating. Tencent Cloud announced today that the two models Hy3 preview and DeepSeek-V4-Pro will end the limited-time free public testing and transition to formal commercial services starting from 10:00 on May 27, 2026, with billing based on model usage. Since March, domestic and foreign cloud vendors have initiated a series of price hikes. Some models of Tencent Cloud have increased by over 400% after switching to token-based billing, while Alibaba Cloud's AI computing power products have increased by 5% to 34%. It is worth noting that the quarterly reports of leading internet companies have solidified the narrative of fundamental reversal. Alibaba has officially entered the "AI commercialization return cycle," with AI-related revenue achieving triple-digit growth for 11 consecutive quarters, and it is expected that the annual recurring revenue from AI models and application services will exceed 10 billion yuan in June, tripling to over 30 billion yuan by the end of the year. Leading internet companies are in a critical breakthrough period for transforming into "AI + Cloud" technology platforms, which is expected to open up profit growth space and usher in value re-evaluation. Focus on the value re-evaluation of Hong Kong stock internet leaders amid the token craze. The Hong Kong stock internet ETF Hwabao (513770) and its linked funds (Class A 017125; Class C 017126) passively track the CSI Hong Kong Connect Internet Index, with the top ten weighted stocks gathering technology giants such as Alibaba-W, Tencent Holdings, and various AI application companies, showcasing significant leading advantages, with T+0 trading and good liquidity. Optimistic about Hong Kong tech but want to reduce volatility? You can also pay attention to the first in the entire market — Hong Kong Large Cap 30 ETF Hwabao (520560), which carries a "Technology + Dividend" dumbbell strategy, with heavy holdings in both Alibaba and other high-elasticity tech stocks, as well as banks, insurance, and other stable high-dividend stocks, making it an ideal foundational tool for long-term allocation in Hong Kong stocks Institutional viewpoint reference sources: ① Bank of China Securities 2026-04-24 "AI Edge Side Welcomes New Logic of Model Catalysis"; ② Soochow Securities 2026-05-16 "After High Volatility, the Logic of the Technology Main Line Remains Unchanged"; ③ Galaxy Securities 2026-05-19 "Collaborative Computing and Power Opens a New Chapter, Green Power Value Faces Revaluation - Token Economy Research Series on the Power Industry"; ④ Xingzheng Strategy 2026-05-07 "AI Diffusion, Don't Forget Hengke". Risk Warning: The Science and Technology Innovation Artificial Intelligence ETF Hwabao and its connected funds passively track the SSE STAR Artificial Intelligence Index, with a base date of December 30, 2022, and published on July 25, 2024; the Power ETF Hwabao and its connected funds passively track the Hwabao WP CSI All Share Electric Utilities Index, with a base date of December 31, 2004, and published on July 15, 2013; the Hong Kong Stock Internet ETF Hwabao and its connected funds passively track the Guotai CSI Hong Kong Connect Internet Index, with a base date of December 30, 2016, and published on January 11, 2021. The composition of index constituent stocks is adjusted in a timely manner according to the index compilation rules, and their historical back-tested performance does not indicate future performance of the index. The individual stocks mentioned in this article are only an objective display of index constituent stocks and do not constitute any stock recommendations, nor do they represent the views of the fund manager or the fund's investment direction. Any information appearing in this article (including but not limited to individual stocks, comments, forecasts, charts, indicators, theories, any form of expression, etc.) is for reference only, and investors must be responsible for any investment decisions made independently. Furthermore, any opinions, analyses, and forecasts in this article do not constitute any form of investment advice to the reader, and our company is not liable for any direct or indirect losses arising from the use of this content. Investors should carefully read the "Fund Contract," "Prospectus," "Fund Product Information Summary," and other legal documents of the fund to understand the risk-return characteristics of the fund and choose products that match their risk tolerance. Past performance of the fund does not indicate its future performance, and the performance of other funds managed by the fund manager does not guarantee the performance of the fund. According to the fund manager's assessment, the Power ETF Hwabao (159146) has a risk level of R3 - medium risk, suitable for balanced (C3) and above investors; the Science and Technology Innovation Artificial Intelligence ETF Hwabao (589520) and the Hong Kong Stock Internet ETF Hwabao (513770) have a risk level of R4 - medium to high risk, suitable for appropriateness ratings of C4 and above investors, and appropriateness matching opinions should be based on the sales institution. Sales institutions (including direct sales institutions of the fund manager and other sales institutions) conduct risk assessments of the above funds according to relevant laws and regulations, and investors should pay attention to the appropriateness opinions issued by the fund manager in a timely manner. The appropriateness opinions from various sales institutions may not necessarily be consistent, and the risk level evaluation results of fund products issued by fund sales institutions must not be lower than the risk level evaluation results made by the fund manager. There may be differences in the fund risk-return characteristics and fund risk levels in the fund contract due to different considerations. Investors should understand the risk-return situation of the fund and carefully choose fund products based on their investment objectives, time horizon, investment experience, and risk tolerance, and bear the risks themselves The registration of the above funds by the China Securities Regulatory Commission does not indicate any substantive judgment or guarantee regarding the investment value, market prospects, and returns of the funds. Fund investments should be made with caution