AI Daily Call Volume Surges 1000x, Huaxia Artificial Intelligence ETF Rises 1.14%

NightAirdropper · 2026-03-25T12:56:32+00:00

As of March 24, the Shanghai and Shenzhen markets surged broadly, with AI-related ETFs and individual stocks performing well. Data from the National Data Administration shows rapid development of China's AI sector, with a significant increase in API call volumes. Analysts believe that growing power consumption and AI hardware applications will have important impacts on the market. Large language models and AI applications are accelerating their iterations, demonstrating coordinated advancement in hardware and content.

NightAirdropper

2026-03-25 12:56:32

Abstract generation in progress

As of 14:55 on March 24, the Shanghai Composite Index rose 1.73%, the Shenzhen Component Index increased 1.37%, and the ChiNext Index gained 0.39%. In the ETF sector, Artificial Intelligence ETF Huaxia (515070) rose 1.14%, with constituent stocks Xin Yuan Shares (688521.SH), Cambricon (688256.SH), and Shenzhen Sanda A up over 5%. Inspur Information (000977.SZ), Donghua Software (002065.SZ), Rockchip (603893.SH), Tongfang Co. (600100.SH), Hengxuan Technology (688608.SH), Goodix Technology (603160.SH), and Heertai (002402.SZ) also increased.

Regarding news, Liu Liehong, Director of the National Data Bureau, stated at a State Council Information Office press conference that by March this year, China’s daily token call volume exceeded 140 trillion, a more than 1,000-fold increase from 100 billion at the beginning of 2024. This fully demonstrates that China’s artificial intelligence development has entered a rapid growth phase, with AI application scenarios continuously deepening, evolving from conversational agents to decision-making and execution intelligent entities, significantly enhancing industry competitiveness.

Guotai Haitong Securities indicated that for power systems, the real pressure and value revaluation are not only driven by annual electricity consumption growth but also by the sustained operation and high load factor of the average total load. Under a neutral scenario, by 2030, data center loads are expected to increase nearly 50 GW compared to 2024, with the additional capacity reaching a level that could substantially impact supply and demand patterns, stability power utilization hours, and system regulation capabilities in multiple key regions.

Tianfeng Securities stated that the iteration speed of large models is accelerating, and AI hardware and content deployment are speeding up. On the model front, Deepseek-v4 is expected to go live soon, with significant upgrades in native multimodal and agent capabilities, and deep optimization for domestic chips, improving software-hardware synergy efficiency. On the application side, Tomato has launched a grayscale test of AI real-person interaction dramas, accelerating the deployment of multimodal interaction capabilities in content scenarios; the Qianwen project plans to release its first AI glasses at MWC 2026, with major companies extending from model competition to AI entry points and hardware ecosystems.

View Original

This page may contain third-party content, which is provided for information purposes only (not representations/warranties) and should not be considered as an endorsement of its views by Gate, nor as financial or professional advice. See Disclaimer for details.

2 Likes