<img src="https://imageproxy.pbkrs.com/https://inews.gtimg.com/om_bt/Ohzai1fs3Got5jty74gXyoOR-wX6rlTX7Hoq3KIY676C0AA/641?x-oss-process=image/auto-orient,1/interlace,1/resize,w_1440,h_1440/quality,q_95/format,jpg" alt="" original-src="https://imageproxy.pbkrs.com/https://inews.gtimg.com/om_bt/Ohzai1fs3Got5jty74gXyoOR-wX6rlTX7Hoq3KIY676C0AA/641"/>
Recently, due to the escalation of geopolitical conflicts, market risk appetite has continued to decline. Coupled with a shift in some funds towards cyclical sectors, the technology sector has faced pressure and volatility. On March 18, the computing power industry chain experienced a collective outbreak. On March 19, under pressure, computing power leasing strengthened against the market trend. On March 20, the computing power sector continued its strength, with the largest index communication ETF ChinaAMC (515050) rising over 4% during trading, and the lowest fee ChiNext Artificial Intelligence ETF ChinaAMC (159381) rising over 3% during trading. Yuanjie Technology (688498.SH) reached a new historical high, while stocks such as Zhuosheng Micro (300782.SZ), Guangku Technology (300620.SZ), Xinyi Sheng (300502.SZ), Zhongji Xuchuang (300308.SZ), and Tianfu Communication (300394.SZ) also strengthened.【ETF secondary market gains do not represent fund net value performance, entering the market involves risks, and investment should be cautious.】
In the past 5 days, the communication ETF ChinaAMC (515050) has received a net inflow of over 160 million yuan.
After the rapid rebound in computing power, should one take profits or continue to hold? Is it a good time to enter, or will chasing high prices lead to being trapped? Let's discuss this in detail today.
What are the background and catalysts for the rebound?
This round of rebound is not merely short-term speculative trading but is catalyzed by better-than-expected industry fundamentals.
On one hand, the domestic computing power sector has initiated a price increase wave, confirming the industry's high prosperity: Alibaba Cloud announced a maximum price increase of 34% for AI computing power, storage, and other products, mainly due to a surge in Token usage. Alibaba Cloud's MaaS business achieved its highest growth rate in history from January to March this year. Baidu (BIDU.US) also released a price adjustment announcement for AI computing power, storage, and other products on the same day. AI computing power-related product services: increased by about 5% to 30%; parallel file storage, etc.: increased by about 30%. These prices will take effect from April 18, 2026.
The popularization of AI applications and the OpenClaw have ignited demand for inference computing power. Coupled with NVIDIA (NVDA.US)'s limited production capacity, rising hardware costs, and gaps in domestic alternatives, the market has entered a &#34;seller's market,&#34; with expectations that short-term price increases are likely to continue.
On the other hand, the simultaneous opening of two major global industry events has released unexpected industrial signals, providing solid fundamental catalysts for the sector's market performance.
Firstly, the OFC 2026 Optical Communication Conference, as the most authoritative technology and industry barometer in the global optical communication field, has core validations of the technological iteration speed and demand prosperity of the optical communication industry: 1.6T optical modules have officially entered the year of mass production, having completed certification from major North American cloud providers, with multiple institutions raising their global shipment expectations for 2026 to over 10 million units, achieving several times growth compared to 2025;
The 3.2T optical module has completed the release and verification of the full-stack technical solution, with the generational iteration pace faster than the market's previous expectations; at the same time, cutting-edge technologies such as CPO, NPO, and OCS have all released commercialization plans, breaking the pessimistic expectations of &#34;peak demand for optical modules and worsening price wars harming industry profitability,&#34; clarifying the medium to long-term growth path of the industry.
Secondly, at the NVIDIA GTC 2026 conference, as a core barometer of the global AI industry, Jensen Huang's keynote speech and technology release set a clear tone for the prosperity of the computing power industry chain. The core industry signals include three points:
First, the demand-side guidance significantly exceeded expectations. Jensen Huang clearly stated that as of now, by 2027, there will be at least $1 trillion in high-confidence computing power-related procurement demand, far exceeding market expectations and directly dispelling concerns about a slowdown in AI computing power demand growth;
<img src="https://imageproxy.pbkrs.com/https://inews.gtimg.com/om_bt/Ohtekm9yP8o98pFDYvzN3-auyYkWW58zmBAjaLDvGNZ98AA/641?x-oss-process=image/auto-orient,1/interlace,1/resize,w_1440,h_1440/quality,q_95/format,jpg" alt="" original-src="https://imageproxy.pbkrs.com/https://inews.gtimg.com/om_bt/Ohtekm9yP8o98pFDYvzN3-auyYkWW58zmBAjaLDvGNZ98AA/641"/>
Second, the industrial phase has clearly shifted. AI has fully transitioned from the era of large model training to the industrial era of inference + intelligent agents + physical AI. Tokens, the smallest semantic units for understanding and generating language in large AI models, can be simply understood as the &#34;language atoms&#34; of the AI world, will become core commodities in the AI era. The inference and intelligent agent scenarios will lead to exponential growth in token consumption, significantly enhancing the continuity and rigidity of computing power demand;
<img src="https://imageproxy.pbkrs.com/https://inews.gtimg.com/om_bt/OO_Iq6lsdWl22WjL2OiJLqBm6ysyMFxpPljWsZ_Z4LOxgAA/641?x-oss-process=image/auto-orient,1/interlace,1/resize,w_1440,h_1440/quality,q_95/format,jpg" alt="" original-src="https://imageproxy.pbkrs.com/https://inews.gtimg.com/om_bt/OO_Iq6lsdWl22WjL2OiJLqBm6ysyMFxpPljWsZ_Z4LOxgAA/641"/>
Third, the technological iteration and interconnection route are clear. The Rubin Ultra architecture GPU, Feynman forward architecture chip, and LPU inference chip integrating Groq technology have been released, achieving a leap in computing power performance and inference efficiency. At the same time, the &#34;copper and optical coexistence&#34; interconnection route has been clarified, stating that the industry chain needs more copper cables, optical chips, and CPO production capacity, calming previous market disputes about the technological route. The leap in computing chip performance will directly drive the upgrade of supporting demand for optical modules, CPO, and other optical interconnection products.
It should be clarified that domestic optical module manufacturers have the capability to participate in the global computing power industry chain, deeply binding with global AI giants such as NVIDIA and Microsoft (MSFT.US). The technological upgrades and capital expenditure expansions of overseas giants will directly translate into orders and performance for domestic leading manufacturers, catalyzing solid fundamental support.
What stage is the computing power sector currently in? Has it accumulated a lot of bubbles?
Here is the conclusion: The computing power sector has officially entered the performance realization period of accelerated prosperity from the early theme speculation stage; the driving logic of the market has shifted from valuation elevation to continuous performance release support, currently in the mid-term of the AI computing power industry's prosperity upward, rather than at the end of the market
Reviewing the three development stages of the computing power market can more clearly anchor the current position:
2023: Anticipation and speculation period, with ChatGPT bringing enlightenment to the AI industry. The core of market pricing is the long-term potential of the AI industry, with the sector showing a general rise characteristic, having a low correlation with orders and performance, and returns mainly coming from valuation increases;
2024-2025: Order verification period, where the market returns to rationality, and the core of pricing shifts to the company's technological barriers and order fulfillment capabilities. The sector shows significant differentiation, with leading companies entering the global giant supply chain continuing to strengthen, while purely thematic stocks gradually decline;
Currently, we have entered the performance fulfillment period and the acceleration of prosperity, with the core feature being that the leading companies in the industrial chain have achieved scaled profit release, and the sustainability and certainty of performance growth have been further validated in the two major events.
From the latest industrial data, optical modules, as the first segment of the computing power industrial chain to fulfill performance, are expected to see leading domestic manufacturers achieve high net profit growth in 2024 and 2025, with growth coming entirely from the simultaneous increase in volume and price of their main business; in 2026, as 1.6T optical modules enter large-scale production, the proportion of high-end products will rapidly increase, and the revenue and profit growth rates of leading manufacturers will remain high, with the gross margin of 1.6T products significantly higher than that of 800G products, and profitability is expected to further improve.
In the computing power leasing segment, the previously feared &#34;overcapacity&#34; has not occurred, and the supply and demand for high-end computing power remain in a tight balance. The core reason is that AI demand has shifted from phase-based training computing power to continuous and rigid reasoning computing power, and the demand for reasoning computing power will grow exponentially with the popularization of AI applications and intelligent agents, providing long-term support for industry prosperity.
Regarding the question of &#34;whether the market has peaked,&#34; it is essential to clarify: the sustainability of stock price increases primarily depends on whether the driving factors are valuation bubbles or performance support. If the valuation is solely driven by sentiment, the market lacks sustainability; however, if corporate performance continues to grow rapidly, even if valuations remain stable, performance growth will drive stock prices up, which is often referred to as &#34;performance digesting valuation.&#34; The current rise in the computing power sector has shifted from early sentiment-driven to fundamentally driven; as long as there is no systemic turning point in industry prosperity and the core logic of performance growth is not disrupted, the market has a sustainable foundation.
How to grasp the computing power price increase market?
For ordinary investors, the computing power industrial chain has rapid technological iterations and significant stock differentiation, facing high non-systematic risks related to technology, orders, and performance falling short of expectations. ETFs, as investment tools for a basket of leading stocks, can both capture the beta returns from the upward trend of industry prosperity and effectively diversify individual stock risks, making them a better choice for ordinary investors to allocate in the computing power sector.
This introduction presents three computing power-related ETF products, which can be matched and selected based on individual investment needs:
Communication ETF ChinaAMC (515050)
Tracks the CSI 5G Communication Theme Index, with holdings covering leading companies in core hardware for computing power such as optical chips, optical modules, optical devices, optical fibers and cables, PCBs, and storage, fully anchoring the core beneficiaries of the CPO/optical communication industrial chain, with constituent stocks mostly being core manufacturers in the global AI computing power supply chain, highly aligned with the core tracks catalyzed by the two major events, and with strong performance fulfillment certainty
Suitable for investors focusing on CPO/optical communication computing hardware high prosperity main line, pursuing precise layout. Off-market connection (Class A: 008086; Class C: 008087)
ChinaAMC SZSE ChiNext Artificial Intelligence ETF (159381)
Tracks the ChiNext Artificial Intelligence Index, with holdings covering the entire AI industry chain including CPO, computing power leasing, AI large models, AI Agents, etc., balancing computing hardware and software applications, highly matching the industry main line of &#34;Inference + Intelligent Body&#34; set by this GTC conference, with outstanding growth elasticity. At the same time, this product is the lowest fee AI-themed ETF in the entire market, with significant cost advantages for long-term holding and regular investment. Suitable for investors optimistic about the long-term development of the entire AI industry, balancing elasticity and holding costs. Off-market connection (Class A: 025505; Class C: 025506)
ChinaAMC Cloud Computing ETF (516630)
Tracks the cloud computing index, focusing on the domestic cloud computing software and hardware ecosystem, with constituent stocks deeply covering AI domestic substitution, DeepSeek, and core AI Agent targets, making it a core tool for laying out the dividends of AI localization. Suitable for investors optimistic about the main line of AI autonomy and controllability, domestic cloud computing, and AI software applications. Off-market connection (Class A: 019868; Class C: 019869).
Risk Warning:
The communication ETF tracks the index of the CSI 5G Communication Theme Index, with index performance for the complete accounting years 2021-2025 as follows: 6.05%, -38.04%, 15.85%, 23.37%, 98.42%. Historical performance of the index does not predict future performance of the fund products.
The ChinaAMC SZSE ChiNext Artificial Intelligence ETF tracks the ChiNext Artificial Intelligence Index, with index performance for the complete accounting years 2021-2025 as follows: 17.57%, -34.52%, 47.83%, 38.44%, 106.35%. Historical performance of the index does not predict future performance of the fund products.
The risk level of the above funds is R4 (medium-high risk), and the specific risk rating results are subject to the ratings provided by the fund manager and sales institutions. This information is for service purposes only and does not constitute any investment advice or commitment, nor does it serve as any legal document. Investors must carefully read the fund's &#34;Fund Contract,&#34; &#34;Prospectus,&#34; and &#34;Product Overview&#34; and other legal documents before investing, fully understanding the risk-return characteristics and product features of this fund, and making rational decisions based on their own risk tolerance. The operation time of funds in our country is relatively short and cannot reflect all stages of stock market development. The market has risks, and entering the market requires caution.
Important Fee Notice: The ETFs mentioned above do not charge subscription fees, redemption fees, or sales service fees. Subscription and redemption agents may charge a commission not exceeding 0.5%, which includes related fees charged by securities exchanges, registration and settlement institutions, etc. In addition, the management fee rate and custody fee rate of the relevant ETFs are deducted from the fund assets, with specific rates shown in the table below
<img src="https://imageproxy.pbkrs.com/https://inews.gtimg.com/om_bt/O7da4K5B7Zx9xgZLjT1lYopHUIKVO6iSXkl--x-zBs5CwAA/641?x-oss-process=image/auto-orient,1/interlace,1/resize,w_1440,h_1440/quality,q_95/format,jpg" alt="" original-src="https://imageproxy.pbkrs.com/https://inews.gtimg.com/om_bt/O7da4K5B7Zx9xgZLjT1lYopHUIKVO6iSXkl--x-zBs5CwAA/641"/>

159381

515050

Recently, the optical communication sector has been under pressure due to geopolitical conflicts and a decrease in market risk appetite, but the computing power industry chain experienced a collective rebound on March 18, with related ETFs such as ChinaAMC Communication ETF and ChinaAMC SZSE ChiNext Artificial Intelligence ETF performing strongly. Several individual stocks reached historical highs, with capital inflows exceeding 160 million yuan. The reasons for the rebound include the domestic computing power price increase wave and better-than-expected industry fundamentals, a surge in AI computing power demand, and market expectations that short-term price increases will continue

Why is optical communication continuously rebounding? Can it still be held?