In-depth Analysis of AI Concept Stocks in October: Divergence Between Computing Power and Application Intensifies, New Logic for Institutional Fund Rebalancing

In early October 2026, AI concept stocks in the A-share market showed clear divergence. As the third-quarter earnings season approaches, performance expectations for leading companies across the AI industry chain vary significantly, with market capital flows showing clear differentiation. This report provides a comprehensive analysis of the current AI concept stock market to help investors grasp the latest investment opportunities.

Overall Market Overview of AI Concept Stocks in October

Since entering October, AI concept stocks have shown a "mixed performance" pattern. Wind data shows that the AI concept index has risen by 3.2% for the month, outperforming the Shanghai Composite Index by 1.5 percentage points, but internal structural differentiation is evident. Among them, computing power-related sectors rose by 5.8% during the month, while application sectors only increased by 1.2%, with the gap expanding to 4.6 percentage points, a new high for the year.

Market analysts point out that this divergence mainly stems from differences in investors' expectations for the commercialization process of different segments of the AI industry chain. As the user base of domestic large models continues to expand, computing power demand is showing explosive growth, while commercialization of applications still faces many challenges, leading to reallocation of capital between different sectors.

Strong Performance of Computing Power Sector, Leading Stocks Rally

The computing power sector has become the absolute protagonist in the AI concept stock market in October. Among them, sub-sectors such as optical modules, AI chips, and liquid cooling heat dissipation have shown particularly outstanding performance. Optcore Technology rose by 18.7% during the month, Zhongji Xuchuang rose by 15.3%, and Cambricon's AI chip business order volume increased by 35% month-on-month, driving its stock price up by 12.6%.

Industry analysis shows that the strength of the computing power sector is mainly driven by three factors: first, the third-quarter AI capital expenditures of major domestic internet companies exceeded expectations, leading to strong demand for optical modules; second, the user base of domestic large models has reached a new high, driving a surge in demand for computing infrastructure; third, policy support continues to increase, with the "East Data West Computing" project Phase II accelerating, providing long-term growth momentum for the computing industry.

Notably, the computing power sector also shows internal differentiation. High-end optical modules, high-performance AI chips and other core segments have seen significantly higher gains than ordinary computing infrastructure, showing the market's preference for core technologies and high-barrier segments.

Application Sector Shows Relatively Weak Performance, B-end Services Become New Hotspot

Compared to the strong performance of the computing power sector, the AI application sector showed relatively weak performance overall in October. Consumer-level AI applications such as AI assistants and AI content creation only rose slightly by 0.8% during the month, while enterprise-level AI services rose counter-trend by 2.3%, becoming the only bright spot in the application sector.

In the enterprise-level AI service field, vertical industry solution providers performed prominently. Enterprise AI service providers such as Yonyou Network and Kingdee International rose by 7.5% and 6.8% respectively during the month, showing that the acceptance of AI services in the B-end market is increasing.

Market analysts analyze that the weak performance of the application sector is mainly constrained by two factors: first, the commercialization path of consumer-level AI applications is not yet clear, and profit models still need to be explored; second, enterprise customers have long procurement decision cycles for AI services, making it difficult to form large-scale orders in the short term. However, as AI technology continues to mature, the application sector is expected to see an inflection point in performance in the fourth quarter.

In-depth Analysis of AI Leading Stocks: Increasing Performance Divergence

As a barometer of AI concept stocks, the performance of leading stocks has an important impact on the entire sector. Since October, leading stocks across the AI industry chain have shown clear differentiation, reflecting the market's differentiated judgment on the business prospects of different companies.

In the computing power field, Optcore Technology, with its leading position in high-end optical modules, saw its third-quarter order volume increase by 62% year-on-year, with gross margin rising to 38%, showing strong stock performance. Cambricon benefited from the acceleration of domestic AI chip substitution, with expected third-quarter profits of 120-150 million yuan, a year-on-year increase of over 300%, with market expectations of continued high growth in the fourth quarter.

In the application sector, iFlytek's third-quarter AI business revenue increased by 45% year-on-year, but gross margin fell to 42%, reflecting intensified market competition. Baidu's ERNIE Bot user base exceeded 200 million, with accelerated commercialization, driving its stock price up by 8.3% during the month.

Notably, some AI concept leading stocks have adjusted due to earnings below expectations. For example, an AI security company's third-quarter AI business growth slowed to 15%, below market expectations, with its stock price falling by 7.2% during the month. This indicates that the market's evaluation criteria for AI companies are shifting from "concept speculation" to "performance realization," and valuation logic is being reshaped.

Analysis of Sector Hotspots: Three Themes Lead the Market

Since October, AI concept stocks have shown significant sector movements, with three themes becoming the focus of market capital: domestic large models, AI computing infrastructure, and enterprise-level AI services.

In terms of domestic large models, as the user base of domestic large models such as MiniMax and ERNIE Bot reaches new highs, related concept stocks have collectively strengthened. Among them, MiniMax concept stocks rose by an average of 12.5% during the month, significantly outperforming the overall AI concept stock performance.

In terms of AI computing infrastructure, liquid cooling technology has become a new hotspot. As the power consumption of AI servers continues to increase, demand for liquid cooling solutions has surged, with related companies' third-quarter order volume increasing by 80% year-on-year, showing outstanding stock performance.

In terms of enterprise-level AI services, vertical industry AI solution providers are receiving capital attention. Especially AI service providers in high-value industries such as finance, healthcare, and manufacturing have seen a year-on-year increase of over 50% in new customer numbers in the third quarter, showing that the enterprise-level AI service market is accelerating penetration.

Institutional Capital Flow Analysis: Clear Signs of Portfolio Rebalancing

Capital flow data shows that since October, AI concept stocks have shown obvious "portfolio rebalancing" characteristics. Overall, AI concept stocks have seen net inflows of about 8.5 billion yuan from main forces, but internal structural differentiation is evident.

From a sector perspective, the computing power sector has received net inflows of 6.2 billion yuan from main forces, while the application sector only received net inflows of 2.3 billion yuan, with a ratio close to 3:1, showing the market's preference for the computing power sector. From an individual stock perspective, leading stocks in core segments such as optical modules and AI chips have received capital attention, while some purely concept-speculative AI application stocks have seen capital reductions.

Public fund rebalancing data shows that in the third quarter, AI theme funds' heavy positions accounted for 42% in computing power-related stocks, an increase of 8 percentage points from the second quarter; application sector stocks accounted for 35%, a decrease of 5 percentage points from the second quarter. This allocation change reflects that institutional investors' judgments on the business prospects of different segments of the AI industry chain are changing.

Post-Market Investment Strategy Recommendations: Grasping Opportunities from Dual-Track Divergence

Based on the current performance of the AI concept stock market, we believe investors can focus on the following investment opportunities:

\n
• Computing Power Track: Focus on leading stocks in core segments such as high-end optical modules, AI chips, and liquid cooling heat dissipation. As the user base of domestic large models continues to expand, computing power demand is expected to maintain high growth, with relatively strong earnings certainty for related companies.
\n
• Application Track: Pay attention to enterprise-level AI service providers, especially leading companies in vertical industry solutions such as finance, healthcare, and manufacturing. As AI technology continues to mature, the enterprise-level AI service market is expected to see an inflection point in performance in the fourth quarter.
\n
• Thematic Investment: Focus on thematic investment opportunities such as domestic large models and AI+industry applications. As policy support increases, related fields are expected to see catalysts.
\n

In terms of risk warnings, attention needs to be paid to risks such as AI technology development not meeting expectations, slow commercialization process, and intensified industry competition. Investors should reasonably allocate AI concept stocks according to their own risk tolerance and avoid chasing high prices and speculation.

Conclusion

The AI concept stock market in October shows a clear divergence trend, with intensified divergence between computing power and application dual tracks. As the third-quarter earnings season approaches, the market's evaluation criteria for AI companies are shifting from "concept speculation" to "performance realization," and valuation logic is being reshaped. Investors should focus on core segments of computing power and enterprise-level AI service providers, grasp structural opportunities, while being alert to the risks of purely concept-based speculation. In the long run, as AI technology continues to mature and commercialization accelerates, all segments of the AI industry chain are expected to see continuous growth opportunities.