According to the latest report from TrendForce, the global market for AI-specific optical transceiver modules is experiencing rapid expansion. The market size is projected to surge from $16.5 billion in 2025 to $26 billion in 2026, representing an annual growth rate exceeding 57%.
In market movements, during the morning session on April 22, the CPO (Co-Packaged Optics) concept for optical modules strengthened again. Tianfu Communication rebounded sharply by over 6% after a significant drop the previous day. The ChiNext AI ETF HuaBao (159363), which has an optical module weighting exceeding 45%, rose over 2% intraday, approaching its previous high and aiming for a fifth consecutive positive day. Real-time turnover exceeded 600 million yuan, ranking first among all AI-themed ETFs in the market by trading activity.
**Hotspot Tracking** Recent analysis from industry research firm TrendForce indicates the global AI-specific optical transceiver module market is entering a phase of high-speed growth. The market size is expected to jump significantly from $16.5 billion in 2025 to $26 billion in 2026, with an annual growth rate surpassing 57%. This notable increase is driven not only by continuous technological upgrades but also reflects a structural reorganization within the optical communication supply chain amid the accelerated construction of AI data centers globally. The explosive growth in AI inference demand is pushing computing architecture away from general-purpose GPUs towards specialized ASICs. The latter demonstrates superior performance in inference scenarios due to deeply optimized energy efficiency and cost advantages, with leading players like Broadcom and Marvell showing strong growth in their custom chip businesses. The core logic behind the shift to ASICs is deploying more computing nodes at lower cost, which increases the total number of endpoints requiring interconnection, thereby providing long-term growth momentum for the optical module and PCB industry chains. Concurrently, giants like NVIDIA are focusing on higher-density chip interconnects and high-speed data transmission in next-generation AI computing architectures, further cementing the central role of optical interconnect technology in replacing traditional copper cable solutions.
**Hotspot Analysis** The global AI-specific optical transceiver module market is witnessing explosive growth. The market size is forecast to reach $16.5 billion in 2025 and expand to $26 billion in 2026, with an annual growth rate exceeding 57%. This growth is primarily fueled by the large-scale development of AI data centers, where high-speed interconnection technology has become a key factor in enhancing computing efficiency. As data centers evolve towards larger scale and centralization, demand for 800G and higher-speed optical modules is surging. Global shipments of 800G+ optical transceiver modules are projected to reach 24 million units in 2025 and jump to nearly 63 million units in 2026, an increase of 260%. However, the substantial demand for high-speed optical modules has created a severe supply bottleneck for laser light sources at the very upstream end of the supply chain. Notably, NVIDIA's strategic monopolization of EML laser chip capacity has led to delivery schedules extending beyond 2027. This trend underscores the critical importance of optical module manufacturing equipment—encompassing packaging, coupling, mounting, and testing—which directly determines production yield and mass production efficiency.
**Investment Rationale** The AI-specific optical transceiver module market is experiencing explosive growth. TrendForce predicts the global market size will leap from $16.5 billion in 2025 to $26 billion in 2026, with a potential annual growth rate exceeding 57%, fully demonstrating the sector's immense potential. As data centers develop towards larger scale and centralization, high-speed interconnection technology becomes crucial for determining the performance ceiling of AI data centers. Global shipments of 800G+ optical transceiver modules are expected to hit 24 million units in 2025 and surge to nearly 63 million units in 2026, a 260% increase. Simultaneously, the explosive growth in AI inference demand is driving a shift in computing architecture towards specialized ASICs, generating more need for interconnecting computing nodes and injecting long-term growth momentum into the optical module industry chain. Global tech giants continue to increase capital expenditures significantly. In the third quarter of 2025, capital expenditures for Microsoft, Google, Meta, and Amazon grew year-over-year by 53%, 83%, 133%, and 55% respectively, highlighting the strong momentum in AI infrastructure build-out. The optical module equipment market is expanding rapidly, with the global market for optical module packaging and testing equipment showing a compound annual growth rate of 71.8%. The 800G optical module equipment market is experiencing explosive growth, presenting significant development opportunities for domestic equipment manufacturers. To capture opportunities in optical modules, CPO, and AI applications efficiently, focus is recommended on the ChiNext AI ETF HuaBao (159363)—noted for its leading scale and liquidity in its category—and its corresponding off-exchange share classes (Class A: 023407, Class C: 023408). The underlying index allocates approximately 70% of its weight to computing power (including leaders in optical modules/CPO), providing comprehensive coverage, and about 30% to AI applications, making it representative of both core computing power and AI application segments.
**ETF Fund Fee Information:** When investors subscribe for or redeem fund shares, the subscription/redemption agency may charge a commission of up to 0.5%. Intra-market trading fees are subject to the rates charged by the securities firm; no sales service fee is levied.
**Connecting Fund Fee Information:** The ChiNext AI ETF Feeder Fund Class C does not charge a subscription fee. A redemption fee of 1.5% applies for holdings under 7 days; holdings of 7 days or more have a 0% redemption fee. A sales service fee of 0.3% applies. For the ChiNext AI ETF Feeder Fund Class A, the subscription fee is 1% for amounts below 1 million yuan, 0.6% for 1 million (inclusive) to 2 million yuan, and a flat 1,000 yuan per transaction for 2 million yuan (inclusive) or above. A redemption fee of 1.5% applies for holdings under 7 days; holdings of 7 days or more have a 0% redemption fee. No sales service fee is charged.
**Risk Warning:** The ChiNext AI ETF HuaBao passively tracks the ChiNext Artificial Intelligence Index. The base date for this index is December 28, 2018, and its release date was July 11, 2024. The annual performance (gain/loss %) of the ChiNext Artificial Intelligence Index from 2021 to 2025 was as follows: 17.57%, -34.52%, 47.83%, 38.44%, 106.35%. The index constituents are adjusted according to the index methodology rules; its backtested historical performance is not indicative of future index performance. The index constituents mentioned are for illustrative purposes only; descriptions of individual stocks are not investment advice in any form and do not represent the holdings or trading动向 of any fund managed by the management company. The fund manager assesses this fund's risk rating as R4 - Medium-High Risk, suitable for Aggressive (C4) and higher investor types. Suitability matching opinions should be based on the selling institution's assessment. Any information appearing herein is for reference only, and investors are responsible for any independent investment decisions. Furthermore, any views, analysis, or forecasts herein do not constitute investment advice of any kind to the reader, and no liability is accepted for any direct or indirect losses arising from the use of this content. Fund investment carries risks; the past performance of a fund is not indicative of its future performance. The performance of other funds managed by the fund manager does not guarantee the performance of this fund. Invest cautiously.
MACD golden cross signals have formed, and these stocks are performing well.
Comments