``` Beyond the Technology M Peak: Two Golden Rules ```
```
As the AI technology sector undergoes a major adjustment, Guotou Securities' strategy team proposes two "golden rules" tested through actual combat, providing investors with a framework to grasp the rhythm in the industrial wave.
With the disclosure of public fund reports in Q2 2026, institutions have reached a historic high in their concentration on AI technology—TMT holdings exceed 60%, and general AI technology is close to 70%, significantly surpassing the previous group concentration levels of the "Mao Index" and "Ning Combination". Coupled with the recent sharp drop in the AI tech sector, market discussions about the "collapse of AI tech group concentration" have intensified.
Guotou Securities strategy analysts Lin Rongxiong and Peng Jingtao stated in a July 23 report that the current adjustment in AI technology is more about profit-taking after a period of overheating, and the first peak of the M-top, marking the end of the industrial wave, has "most likely not yet arrived". The report also points out that after this decline, the core pricing logic of AI tech will gradually shift from "price hike signals" to "volume growth" signals, and core tracks within may experience another switch.
First Peak of M-top Not Confirmed, All Three Criteria Not Triggered
The Guotou Securities strategy team previously introduced and led market-wide discussions on the "technology sell point" in its reports: "Gray Decisions: The Tech Sell Point", "Tech M-top: First Peak", and "Tech M-top: Second Peak".
The report sets three conditions for confirming the first M-top. First, qualitative trading signals: The crazy stage of the first peak is characterized by the leaders driving second- and third-tier stocks, with the leader’s excess not far behind—however, in this round’s Q2 AI tech, the excess of second- and third-tier stocks is far higher than the leaders, so this signal is unclear. Second, quantitative trading signals: The A-share high-to-low index needs to fall from well above 60%—historically, when the "Mao Index" and "Ning Combination" group truly collapsed, this index was as high as over 90%; as of July 17, the index has rapidly fallen to 33%, near the historical low range, indicating this round’s high-to-low trend is nearing its end. Third, industry logic signals: The leader’s single-quarter earnings growth must reach the highest point—currently, most of the market expects the highest single-quarter earnings growth for AI leaders has not yet arrived.
The report emphasizes that before the end of the industrial trend, and without clear macro "grey rhino" or collapse in industry competitive structure, the market is likely to return to the main trend of industry after a short-term adjustment.
Golden Rule One: Four Stages of Tech Industry Investment, 2026 Should Switch to Supply-Demand Gap
The report’s first golden rule is the four-stage framework for tech growth industry investment: Buy giants (explosive products emerge) → infrastructure (giants begin large capital spending) → key sections of the industry chain (industry chain forms, completes 0-1) → supply-demand gap (the process from 1-100).
The report verifies this using the new energy vehicle industry boom: the giant stage corresponds to Tesla Model 3 becoming a hit; infrastructure stage corresponds to charging stations and grid equipment (focused on pricing since 2019); key industry chain sections corresponds to lithium batteries (pricing focus since 2020); supply-demand gap stage corresponds to upstream lithium resource price hikes and downstream auto parts volume increases in 2021.
Mapped onto this round of AI boom: the giant stage corresponds to Nvidia, Microsoft, etc. upon ChatGPT’s appearance in 2023; infrastructure stage corresponds to computing power and optical modules in 2024; key industry chain sections to AI chips in 2H24; supply-demand gap application end covers storage, power, energy storage, copper, and other upstream supply-demand gaps, as well as multimodal, smart driving, AI large model software, and other downstream supply-demand gaps.
Based on this, the report concludes that AI tech investment should migrate to the fourth stage of supply-demand gap in 2026, summarized as "guarding the big light and buying the aperture"—analogous to the shift in 2021 to lithium mines and auto parts around CATL, now to second- and third-tier companies around AI supply-demand gaps. The report also notes the sharp rise in memory prices since H2 2025 is highly similar to the lithium price surge in 2021.
Golden Rule Two: "Big-Small-Big-Small" Pattern, Rebalancing Between M-tops
The second golden rule is the "big-small-big-small (strong alpha)" pattern based on three stages: trend-group-crazy: Trend stage trades big leaders → group stage trades mid-small companies → crazy stage sees core leaders surge → after first M-top, shift to strong alpha second- and third-tier stocks.
The report supports this with three historical cases.
Consumption upgrade boom (2019-2021): Trend stage led by Kweichow Moutai, Wuliangye, then group expansion to the whole liquor industry chain, entering crazy stage after December 2020 with Moutai leading second- and third-tier stocks to peak and clear valuation ballooning; between M-top one and two, second- and third-tier stocks like Jiuguijiu and Shede performed better than Moutai.
New energy industry boom (2020-2022): Trend started from CATL, group spread to lithium battery chain and auto parts, crazy stage after October 2021 with CATL leading second- and third-tier stocks; between M-top one and two, second- and third-tier stocks like DFD Nano, Tianqi Lithium (volume and price both up) outperformed CATL.
Mobile internet boom (2013-2015): Trend led by Apple, group expansion to the Apple chain and ChiNext board, after March 2015 crazy stage led by Eastmoney, Hundsun & second- and third-tier stocks; from June between M-top one and two, second- and third-tier stocks like Shunwang Tech, Wangsu Tech performed better than the core leaders.
All three historical periods show that after the first M-top, the investment focus should not be clinging to the leader, but should shift portfolio allocation to ascending second- and third-tier stocks.
Q2 Report: Institutions Extreme Group AI Hardware, "Add Tech Reduce Cyclicals" Accelerates
Q2 2026 data reveal the current allocation patterns of institutions. Among secondary industries, active funds have significantly increased positions in semiconductors, components, and communication equipment, while significantly reducing positions in industrial metals, chemical pharmaceuticals, agrochemicals, auto parts, etc. The flow of funds from traditional manufacturing to AI tech remains in acceleration.
At the primary industry level, the electronics industry’s overweight ratio exceeds 20%, standing out; electronics and communications are absolute leaders in both overweight and increased positions, forming the "twin stars" at the core of public fund holdings. Historical quantiles show electronics, communications, machinery, and building materials are all at historical highs.
Within AI tech, institutions are highly concentrated in AI hardware, integrated circuits, PCB tracks; at the tertiary industry level, the most significant increases are in semiconductor equipment and network connections/towers. Among individual stocks, active increases are mostly focused in optical modules, semiconductors, and electronics; New Yi Sheng, Cambricon, SMIC, etc. top the list for increased position value, fully consistent with industry direction, confirming AI compute hardware is the strongest current institutional consensus.
For resource products, nonferrous metals, oil & petrochemicals, and basic chemicals are all underweighted; there is internal differentiation within the overseas supply chain, with white goods, agrochemicals, and electrical equipment reduced, and institutional preference for overseas supply chain converging towards AI industry chain overseas expansion.
The report also notes that among fund types, equity-biased hybrid and flexible allocation funds increased in Q2 overall, ordinary stock funds held steady. The incremental configuration power in the current market mainly comes from flexible products’ added allocation to tech. Worth noting, historical experience shows the first M-top does not correspond to the highest position level, the second M-top does—which means before reaching the second M-top, the AI group concentration could strengthen further.
Risk Warning and DisclaimerThe market is risky, and investment requires caution. This article does not constitute personal investment advice, nor does it take into account the unique investment goals, financial status, or needs of individual users. Users should consider whether the opinions, views, or conclusions in this article suit their situation. Investing based on this article is at your own risk. ```