坚定加仓！创业板人工智能ETF（159363）缩量退守10日线，资金单日扫货超3亿份！博弈年报行情？

Market Overview - The ChiNext AI sector experienced fluctuations with a slight pullback, while funds continued to buy on dips. Most constituent stocks declined, with notable gainers including Hangyu Micro rising over 6% and BlueFocus up over 3%. Conversely, Ruijie Networks led the decline with a drop of over 10% [1][7] - The popular ETF tracking the ChiNext AI sector (159363) saw a decrease of 1.56%, retreating to the 10-day moving average, with a trading volume of 868 million yuan. After a significant net inflow of nearly 1.7 billion yuan over the past week, the fund recorded a net subscription of 322 million units today, indicating continued accumulation [1][7] Investment Strategy - According to Industrial Securities, the focus on performance expectations in the second half of January during the earnings disclosure period suggests a structural adjustment in the market. As market sentiment returns to rationality, earnings will become a key factor driving market movements, with previous hot sectors facing performance validation [3][9] - The technology sector, particularly in upstream computing hardware (communications, semiconductors), has seen upward revisions in profit forecasts since November last year, indicating a potentially positive fundamental support. The communications industry has underperformed in the AI hardware sector since the beginning of the year, suggesting it may attract funds focused on fundamentals during the earnings disclosure period [3][10] Future Outlook - CITIC Securities anticipates that 2026 will be a pivotal year for AI applications, as major cloud service providers (CSPs) have invested heavily in capital expenditures for over three years. The development of AI applications will continue to drive the growth of computing power and model training iterations [4][10] - The ChiNext AI ETF (159363) is positioned to benefit directly from the commercialization of AI technology, with approximately 60% of its portfolio allocated to computing power (primarily optical modules) and 40% to AI applications, representing both core computing and true AI application sectors [4][10]