⚠️ Editorial note: The open source ecosystem in China operates under a distinct institutional framework — characterized by state-led initiatives, intranet-like boundaries, and top-down governance. Readers should be aware that this context differs from the community-driven open source model common in other regions. The term “open source” as used in Chinese media may refer to practices that diverge from the conventional definition.

China Open Source Daily — 2026-10-09

🏗️ Institutional Change — OpenHarmony’s 《生成式AI工具使用与开源贡献策略》 (Generative AI Tool Use and Open Source Contribution Strategy) Published as the First Large-Chinese-OS-Community Official AI-Tool-Use Rulebook For Open-Source Contributions

1. 开放原子开源基金会 (OpenAtom Foundation) / 开源鸿蒙社区 (OpenHarmony Community) (2026-10-08) — 《生成式AI工具使用与开源贡献策略》 officially released

2. OpenHarmony Community WeChat official account — “开源鸿蒙《生成式AI工具使用与开源贡献策略》正式发布” (OpenHarmony 《Generative AI Tool Use and Open Source Contribution Strategy》 Officially Released)

3. Community rulebook — eight sections: (i) Basic Principles (人类主导/责任到人/安全优先/合规必达 — human-led, accountability-to-person, security-first, compliance-mandatory); (ii) Permitted AI-Assistance Scenarios (code-writing, documentation translation, test-case authoring, problem-analysis, patch-explanation — five permitted scenarios); (iii) Disclosure Norms (four categories that must be disclosed + four categories exempt from disclosure + heightened requirements for high-sensitivity modules); (iv) Six Prohibited Behaviors (unverified AI outputs, fabricated test data, batch-issue-PR flooding, uploading sensitive data to AI, sole-reliance on AI for code review, unapproved AI-agent deployment); (v) License / Privacy / Data controls; (vi) Issue / Bug / Security-Vulnerability special rules; (vii) Reviewer-side AI-review rules; (viii) Community-enforcement and policy-iteration mechanism.

4. Rulebook’s sharpest institutional provisions: (a) Maintainers may demand code-contribution submission withdrawal if the contributor cannot explain AI-generated code’s logic, risk, or license provenance; (b) High-severity vulnerabilities require two-person cross-verification — AI-inference is not permitted to determine risk-grade; (c) Undisclosed AI contribution with a security risk triggers PR closure or PR take-back; (d) Four-dimensional governance stack: manual review + full test suite + transparent disclosure + tiered verification.

5. Rulebook’s positioning claim (community channel’s own metadata): “The world’s first generative-AI-tool-use rulebook officially issued by a large operating-system open-source community.”

OpenHarmony’s October 8, 2026 publication of the 《生成式AI工具使用与开源贡献策略》 is the first-documented instance in this series of a large Chinese operating-system open-source community’s official community channel publishing a formal AI-tool-use rulebook for open-source contributions at the AI-governance-rule-execution layer simultaneously with this series’ September 8 briefing’s 14-billion-device-OpenHarmony-base-layer-plus-90-to-70-percent-contribution-share-decline-plus-Hubble-20-percent-Huawei-vendor-arm-stake institutional object and this series’ September 14 briefing’s state-directed-open-source-project-selection-plus-independent-open-source-license-compliance-guidance-plus-community-first-release-plus-research-evaluation-incorporation-plus-quantitative-project-target institutional object — a two-surface-simultaneous AI-governance-rule-execution institutional-form observation.

The sharpest institutional-economics fact of the day: OpenHarmony’s official community channel publishes a formal AI-tool-use rulebook for open-source contributions that names four institutional principles (人类主导 / 责任到人 / 安全优先 / 合规必达 — human-led, accountability-to-person, security-first, compliance-mandatory) as the operative governance frame, and operationalizes three Williamson L3 governance mechanisms: (a) mandatory disclosure with four categories of mandatory-disclosure and four categories of exempt-from-disclosure scenarios; (b) two-person cross-verification for high-severity vulnerabilities, with an explicit prohibition on AI-inference directly determining risk-grade; (c) maintainer-side right to demand code-contribution withdrawal if the contributor cannot explain the logic, risk, or license provenance of AI-generated code. The rulebook is not a technical standard — it is an L3 governance mechanism that operationalizes the community’s own AI-tool-use rule surface as a community-governance instrument rather than a platform-vendor instrument. The OpenHarmony community’s positioning claim that this is “the world’s first generative-AI-tool-use rulebook officially issued by a large operating-system open-source community” is an institutional-positioning claim by the community channel itself, not an external verification — but the claim’s institutional significance is that a large Chinese open-source community is formalizing a rule surface at the community level, at the same institutional layer at which this series’ September 14 briefing documented the state-directed AI+Software-Special-Action-Implementation-Plan at the national-industrial-policy layer.

The AI-assistance-disclosure layer. Per the community rulebook: the four mandatory-disclosure categories are (i) core-logic AI-generated code with a high share; (ii) test suites originating from AI; (iii) large-scale cross-module refactoring performed by AI; (iv) security reports dependent on AI inference. The four exempt-from-disclosure categories are (i) spelling and grammar polish; (ii) single-line IDE autocomplete; (iii) code formatting; (iv) comment polishing. High-sensitivity modules receive heightened requirements — mandatory security self-checklist, mandatory second-verification reviewer, mandatory security test-case list. The institutional reading is that the AI-assistance-disclosure layer is being operationalized at the community-governance-plus-reviewer-side-rules institutional-form layer — the community is being asked to draw a specific boundary between AI-assisted-contributions-that-require-disclosure and AI-assisted-contributions-that-do-not, and the boundary is being drawn at the community level rather than at the platform-vendor level (GitHub’s Copilot policy, Apache’s AI policy) — a structurally distinct institutional-form object from the Western-platform-vendor-instrument layer.

The six-prohibited-behaviors layer. Per the community rulebook: (i) unverified AI outputs; (ii) fabricated test data; (iii) batch-issue-PR flooding; (iv) uploading sensitive data to public AI systems; (v) sole-reliance on AI for code review; (vi) unapproved deployment of AI agents. The institutional reading is that the six-prohibited-behaviors layer is being operationalized at the community-governance-plus-reviewer-side-rules-plus-CI-gate institutional-form layer simultaneously with the AI-assistance-disclosure layer — a two-layer-simultaneous AI-governance institutional-form observation. The prohibition on uploading sensitive data to public AI systems — specifically “未公开内核不上传公有AI” (unpublished kernel is not to be uploaded to public AI) — is a first-documented instance in this series of a large Chinese OS-community placing a public-AI-system-uploading prohibition on unpublished-kernel-code contributions at the community-governance-plus-reviewer-side-rules layer simultaneously with this series’ October 1 briefing’s documentation of the open-source-label-as-market-positioning-plus-Xinchuang-market-instrument layer and this series’ October 5 briefing’s Model-Diplomat-MOFCOM-plus-NDRC-tiered-AI-model-export-control layer.

The two-person-cross-verification-and-prohibition-on-AI-inference-directly-determining-risk-grade layer. Per the community rulebook: “AI仅辅助整理日志Issue必须人工核验禁止AI自动提交Issue” (AI only assists in log organization, issues must be manually verified, AI automatic issue-submission is prohibited); “AI推理不能直接定级风险” (AI inference cannot directly determine risk-grade); “高危漏洞需两人以上交叉验证” (high-severity vulnerabilities require two-person cross-verification); “未公开漏洞不输入公有AI” (undisclosed vulnerabilities must not be input into public AI). The institutional reading is that the two-person-cross-verification-and-prohibition-on-AI-inference-directly-determining-risk-grade layer is being operationalized at the community-governance-plus-reviewer-side-rules institutional-form layer — a structurally distinct institutional-form object from the Western-platform-vendor-instrument layer this series’ prior briefings’ Western-foundation-governance-seat layer (PyTorch Foundation, September 10 briefing; AAIF, September 7 briefing) documented.

The institutional-economics question this cycle’s briefing documents but does not answer. Whether the OpenHarmony community channel’s formalization of a community-level AI-tool-use rulebook converts the OpenHarmony community’s own institutional-form object (this series’ September 8 briefing’s 14-billion-device-plus-90-to-70-percent-contribution-share-decline-plus-Hubble-20-percent-stake-plus-2028-ecosystem-platform-2028-target) into a community-governance-instrument-object rather than a platform-vendor-instrument-object, and whether the same AI-governance-rule-execution surface is being operationalized at the community level simultaneously with the same state-directed-industrial-policy surface this series’ September 14 briefing documented at the national-industrial-policy layer.

Institutional significance: OpenHarmony’s 《生成式AI工具使用与开源贡献策略》 is the first-documented instance in this series of a large Chinese operating-system open-source community’s official community channel being operationalized at the AI-governance-rule-execution layer simultaneously with this series’ September 8 briefing’s OpenHarmony-14-billion-device-base-layer-plus-90-to-70-percent-contribution-share-decline layer and this series’ September 14 briefing’s state-directed-open-source-project-selection-and-community-first-release layer — a two-surface-simultaneous AI-governance-rule-execution institutional-form observation.

Sources:

🏗️ Institutional Unwinding — 国产信创外传 Documents the Xinchuang (State-Informatization) Blind-Replacement Slowdown and the “Business-First Value-Based Rollout” Reframe, Positioning Huawei as the Sole Beneficiary of Xinchuang Stratification

1. 国产信创外传 (Chinese State-Informatization Industry Outlet, WeChat) — “信创厂商自认盲目替换·告别野蛮替换时代” (Xinchuang Vendors Admit Blind Replacement — Farewell to the Barbaric-Replacement Era)

2. 国产信创外传 (WeChat) — “为什么越来越多单位暂停盲目国产化替换·真正的信创落地逻辑变了” (Why Increasingly More Units Are Pausing Blind Domestic-Localization Replacement — The Real Xinchuang Rollout Logic Has Changed)

3. Institutional claim — enterprises, government units, and energy-sector units have slowed one-size-fits-all Xinchuang replacement, retreating from replacement-rate-first KPIs toward business-first, risk-controlled, value-based rollout.

4. Institutional failure-mode diagnosis — the four Xinchuang-era problems: (i) 老旧业务系统适配工作量远超预估 (adaptation workload for legacy business systems far exceeds initial estimates); (ii) 上线后性能不达标 (post-deployment performance does not meet the initial benchmarks); (iii) 小众外设和工控软件缺少驱动 (niche peripheral devices and industrial-control software lack drivers); (iv) 后期运维成本飙升 (long-term operational-maintenance costs spike).

5. Institutional KPI reframe — from “replacement-coverage-rate” (替换覆盖率) to “business safety-driven deployment” (业务安全驱动); new three principles: 分级替代 (tiered replacement — office / management / production-core), 先评估后改造 (assess-before-modify), 灰度落地双轨验证 (gray-rollout dual-track verification); the operative frame is “能替则替、难替缓替” (replace-what-can-be-replaced, defer-what-cannot-be-replaced) — prioritizing business continuity over replacement-rate-maximization.

6. Institutional-risk taxonomy — the three risks being avoided by the pause: 业务中断风险 (business-interruption risk — critical systems in energy and finance cannot tolerate downtime), 隐性适配风险 (hidden-adaptation risk — legacy systems without source-code or vendor support), 长期运维风险 (long-term O&M risk — O&M teams lack the required platform-maintenance experience).

7. Institutional positioning claim — the piece explicitly states that “the goal of Xinchuang has never been to make everything domestic; the goal is critical-business autonomy and controllable risk” (信创的目标从来不是全部国产，而是关键业务自主可控风险可管控).

8. Institutional-industry-dynamics claim — “Huawei is the sole beneficiary in the Xinchuang stratification period; as the high-end vendor with its own chip design and process capability, Huawei is the only beneficiary in the stratification period; low-end markets are squeezed by high-end markets” (华为在信创分化期中是唯一受益的厂商，高端市场挤压低端市场).

The 国产信创外传’s two-piece publication is the first-documented instance in this series of the Xinchuang-blind-replacement-slowdown institutional object being operationalized at the state-informatization-industry-outlet-plus-value-based-rollout-plus-huawei-as-sole-beneficiary institutional-form layer simultaneously with this series’ October 1 briefing’s Higobase-Shandong-SASAC-Guantou-Jiehhe-5-year-exit-cycle-plus-1200-SOE-digital-technology-subsidiary-collapse layer and this series’ October 1 briefing’s open-source-label-as-market-positioning-vs-open-source-institutional-identity distinction layer — a two-surface-simultaneous Xinchuang-institutional-unwinding institutional-form observation.

The sharpest institutional-economics fact of the day: the 国产信创外传 piece’s own operational claim that “the goal of Xinchuang has never been to make everything domestic; the goal is critical-business autonomy and controllable risk” (信创的目标从来不是全部国产，而是关键业务自主可控风险可管控) — paired with the piece’s operational framing of “replace-what-can-be-replaced, defer-what-cannot-be-replaced” (能替则替、难替缓替) — is the sharpest documented retreat from the Xinchuang-era replacement-rate-first-KPI layer to the business-first-value-based-rollout layer. The retreat is not described as a policy change; the piece states that policy has not changed (政策方向没变). The retreat is described as an operational reframe — 从指标驱动转向业务安全驱动 (from KPI-driven to business-safety-driven) — occurring at the enterprise-deployment layer in response to the piece’s own four documented institutional failure modes. The piece’s institutional-positioning claim — “the goal of Xinchuang has never been to make everything domestic” — is a first-documented instance in this series of a state-informatization-industry outlet publicly reframing the operational objective of the same Xinchuang institutional object this series’ October 1 briefing documented at the Xinchuang-tail-mid-tier-domestic-replacement-database-vendor-exit layer and this series’ October 1 briefing’s 1200-SOE-digital-technology-subsidiary-collapse layer — the state-informatization-industry-outlet is now publishing a value-based-rollout reframe at the Xinchuang-institutional-object layer, at the same institutional layer at which the state capital is being rotated away from the Xinchuang-tail-mid-tier vendor toward the frontier-AI-consolidation-head.

The three-Xinchuang-era-failure-mode layer. Per the piece: “老旧业务系统适配工作量远超预估，上线后性能不达标，小众外设和工控软件缺少驱动，业务流程卡壳，一次性全量割接风险高，一旦故障直接影响核心生产，后期运维成本飙升，团队缺乏国产平台运维经验” (Legacy business-system adaptation workload far exceeds initial estimates; post-deployment performance does not meet initial benchmarks; niche peripherals and industrial-control software lack drivers; business flows are stalled; one-shot full-scale cutover carries high risk, and a failure directly affects core production; long-term O&M costs spike; O&M teams lack domestic-platform operational-maintenance experience). The institutional reading is that the three-Xinchuang-era-failure-mode layer is being operationalized at the state-informatization-industry-outlet-plus-value-based-rollout institutional-form layer — the piece is not just reporting a slowdown; it is publishing a specific four-item failure-mode diagnosis that pairs an operational-technical diagnosis (adaptation workload, performance, driver coverage) with an institutional-structural diagnosis (one-shot cutover risk, long-term O&M cost, team experience gap). This is a structurally distinct institutional-form object from the Western-market-competitive-positioning diagnostic layer.

The business-first-three-principles layer. Per the piece: (i) 分级替代 (tiered replacement — non-core office systems, management systems, and production-core business systems are treated as three distinct tiers); (ii) 先评估后改造 (assess-before-modify — no hardware procurement before a full current-state inventory, system-dependency mapping, peripheral-peak-pressure assessment, and modification-workload-and-risk estimation); (iii) 灰度落地双轨验证 (gray-rollout dual-track verification — simulation-environment testing first, then dual-track parallel with gray cutover, with a complete rollback plan). The institutional reading is that the business-first-three-principles layer is being operationalized at the state-informatization-industry-outlet-plus-value-based-rollout institutional-form layer — a structurally distinct institutional-form object from the state-directed-industrial-policy-instrument layer this series’ September 14 briefing documented. The Xinchuang-institutional-object is being reframed at the state-informatization-industry-outlet layer as a business-first, risk-controlled value-based-rollout instrument rather than a state-directed-replacement-rate-instrument.

The “Huawei-as-sole-beneficiary-in-Xinchuang-stratification-period” layer. Per the piece: “华为在信创分化期中是唯一受益的厂商，高端市场挤压低端市场，华为作为高端厂商自有工艺厂商是分化期中唯一受益者” (Huawei is the sole beneficiary in the Xinchuang stratification period; high-end markets squeeze low-end markets; Huawei, as a high-end vendor with its own chip design and process capability, is the sole beneficiary in the stratification period). The piece’s own diagnostic — 华为作为高端厂商自有工艺厂商 (Huawei as a high-end vendor with its own process capability) — is a first-documented instance in this series of the Xinchuang-institutional-object’s stratification being operationalized at the high-end-vendor-only-beneficiary-plus-low-end-market-squeeze layer simultaneously with this series’ October 1 briefing’s Xinchuang-tail-mid-tier-vendor-exit layer — a two-layer-simultaneous Xinchuang-stratification institutional-form observation. The institutional reading is that the Xinchuang-institutional-object is being stratified into a head-and-tail configuration with a specific operational outcome: the state capital is being rotated toward the frontier-AI-consolidation-head (this series’ September 1 briefing’s National AI Industry Investment Fund’s $8.8B-in-capital consolidation of Moonshot at $35B and DeepSeek at $74B), while the Xinchuang-tail-mid-tier vendors (this series’ October 1 briefing’s Higobase-plus-1200-SOE-digital-technology-subsidiary-collapse layer) are being exited, and the head of the Xinchuang stratum (Huawei as a high-end vendor with its own chip design and process capability) is being positioned as the sole survivor — the Xinchuang-institutional-object’s own stratification being operationalized at three institutionally distinct layers simultaneously.

The institutional-object pairing with this series’ October 1 briefing’s Xinchuang-tail-mid-tier-vendor-exit layer. This series’ October 1 briefing documented the Higobase Shandong-SASAC-Guantou-Jiehhe-5-year-exit-cycle as the Xinchuang-tail-mid-tier-vendor-exit institutional-form object, paired with the 1200-SOE-digital-technology-subsidiary-collapse as the same state-capital-plus-SOE-digital-technology institutional architecture’s unwinding side. This cycle’s briefing documents the same Xinchuang-institutional-object at the head-of-stratification-plus-huawei-as-sole-beneficiary-plus-value-based-rollout-reframe layer — the Xinchuang-institutional-object’s unwinding side and the Xinchuang-institutional-object’s head-of-stratification side are being documented at the same institutional-object layer, one side at the tail-vendor-exit-plus-SOE-subsidiary-collapse layer, the other side at the high-end-vendor-sole-beneficiary-plus-value-based-rollout-reframe layer.

The institutional-economics question this cycle’s briefing documents but does not answer. Whether the 国产信创外传’s “business-first, value-based rollout” reframe converts the Xinchuang-institutional-object from a state-directed-replacement-rate-instrument to a business-first-value-based-rollout-instrument at the state-informatization-industry-outlet layer simultaneously with the same state-capital-plus-SOE-digital-technology institutional architecture’s unwinding side this series’ October 1 briefing documented at the Xinchuang-tail-mid-tier-vendor-exit layer, and whether the “Huawei-as-sole-beneficiary-in-Xinchuang-stratification-period” positioning converts the Xinchuang-institutional-object’s head-and-tail configuration into a specific operational outcome (the state capital being rotated toward the frontier-AI-consolidation-head while the Xinchuang-tail-mid-tier vendors are exited and the high-end-vendor-only-beneficiary is positioned as the sole survivor) at the same institutional-object layer.

Institutional significance: The 国产信创外传’s two-piece Xinchuang-blind-replacement-slowdown-plus-business-first-value-based-rollout-plus-huawei-as-sole-beneficiary publication is the first-documented instance in this series of the Xinchuang-blind-replacement-slowdown institutional object being operationalized at the state-informatization-industry-outlet-plus-value-based-rollout-plus-huawei-as-sole-beneficiary institutional-form layer simultaneously with this series’ October 1 briefing’s Xinchuang-tail-mid-tier-vendor-exit-and-1200-SOE-digital-technology-subsidiary-collapse layer — a two-surface-simultaneous Xinchuang-institutional-unwinding institutional-form observation.

Sources:

🏗️ Institutional Change — Richard Yu’s September 29, 2026 International-Media Roundtable Documents Huawei’s “Essentially Broken Free from US Tech Dependency” vs. China’s “Still Relies on DUV” Dichotomy, With the Autonomous-Controllable-Fresh-News Outlet’s Selective-Headline-Framing as a First-Documented Instance of the Autonomous-Controllable-Narrative’s Market-Media-Plus-Administrative-Narrative Selective-Transmission Mechanism

1. 自主可控新鲜事 (Autonomous-Controllable Fresh News, WeChat) — “余承东谈芯片鸿蒙出海光刻机半导体产能·华为已基本摆脱对美国技术依赖但中国仍依靠DUV”

2. Richard Yu (Huawei常务董事/Investment Review Committee Chairman/Consumer BG Chairman), 2026-09-29 international-media roundtable on Huawei Consumer BG and HarmonyOS ecosystem.

3. Cumulative R&D investment: 1.8 trillion yuan over 2007–2025; R&D ratio raised from 10% to above 20% over the same period.

4. Huawei’s workforce: 200,000 employees, nearly half in R&D.

5. OpenHarmony device count: 90 million+ devices, expected to break 100 million by year-end; application-and-service ecosystem: 450,000+; global-registered-developer count: 11 million+; Chinese-foldable-phone market share: 75%; Mate XT 2 has shipped approximately 100,000 units in the first 15 days after launch.

6. Chip performance: 麒麟 (Kirin) 9050 Pro whole-machine performance improvement: 31%, using the LogicFolding chip architecture.

7. Huawei’s AI product line: 昇腾 (Ascend) + openPangu + 鸿蒙小艺 (HarmonyOS XiaoYi).

8. The 2025-mate-90-series launch: 华为手机海外销量只有几百万台 (Huawei smartphone overseas sales are only a few million units); the overseas-market expansion plan is characterized as “初歩计划，一步步逐渐推进” (a preliminary plan, being advanced step-by-step gradually).

9. Component-cost pressure: memory-price increases are adding over US$200 per phone in cost; Huawei’s own price-increase velocity is “very slow”, and overall profit margin is declining rapidly.

10. The sharpest structural fact of the day — the internal-tension claim inside the same Yu statement: “华为已基本摆脱对美国技术的依赖” (Huawei has essentially broken free from its reliance on US technology) and “中国目前仍依靠DUV” (China still relies on DUV lithography).

11. Yu’s own definition of LogicFolding: “不是EUV替代品而是用面积换工艺的架构方法” — not an EUV substitute, but an architecture method that trades chip area for process capability under restricted access to advanced processes.

12. 自主可控新鲜事’s own headline-framing (per the WeChat input metadata): “选择性强调’基本摆脱’（华为公司层面）弱化了’仍依靠DUV’（中国产业层面）” — selectively emphasizes “essentially broken free” (Huawei-company layer) and weakens “still relies on DUV” (Chinese-industry layer).

Richard Yu’s September 29, 2026 international-media roundtable, as re-published by the 自主可控新鲜事 WeChat piece, is the first-documented instance in this series of the autonomous-controllable-narrative-plus-market-media-selective-transmission institutional object being operationalized at the individual-vendor-vs-national-industry-dichotomy layer simultaneously with this series’ October 8 briefing’s Huawei-Ascend-CANN-external-contributor-majority layer, this series’ September 30 briefing’s Huawei-Ascend-Tribe-openPangu-2.0-Training-plus-openPangu-2.0-RL-plus-Ascend-910B-NPU-only-weights-release layer, and this series’ September 18 briefing’s HUAWEI CONNECT CANN-community-open-source-development layer — a three-surface-simultaneous Huawei-narrative-plus-vendor-institutional-form-plus-autonomous-controllable-narrative-structural-fact institutional-form observation.

The sharpest institutional-economics fact of the day: the internal tension inside the same Yu statement — “Huawei has essentially broken free from its reliance on US technology” and “China still relies on DUV lithography.” The 自主可控新鲜事 piece’s own characterization of the internal tension — “这不是矛盾，而是自主可控叙事的核心结构” (This is not a contradiction; it is the core structure of the autonomous-controllable narrative) — is the sharpest institutional-economics observation of the day, and it is not the piece’s author’s own characterization — it is the piece’s author’s characterization of the vendor-official statement’s own internal structure. The institutional reading is that the individual-vendor-vs-national-industry dichotomy is being operationalized at the autonomous-controllable-narrative-plus-market-media-selective-transmission institutional-form layer — the vendor (Huawei) can claim a company-level autonomy-through-system-engineering claim (LogicFolding as an architecture-trades-area-for-process method), while the national industry level (China’s chip-manufacturing base) still relies on DUV lithography for its advanced chip-manufacturing base. The individual-vendor-vs-national-industry dichotomy is being operationalized at the vendor-official-plus-industry-structural-fact layer simultaneously with the autonomous-controllable-narrative-selective-transmission layer.

The LogicFolding-not-EUV-substitute layer. Per Yu: “LogicFolding是用部分技术获取受到限制情况下提升自身能力的一种方法，不是EUV替代品，而是用面积换工艺的架构方法” (LogicFolding is a method for enhancing one’s own capability under conditions where access to some technologies is restricted; it is not an EUV substitute; it is an architecture method that trades chip area for process capability). The institutional reading is that the LogicFolding-not-EUV-substitute layer is being operationalized at the vendor-official-plus-industry-structural-fact-plus-chip-architecture-trading-method layer — the vendor is characterizing its chip architecture as an area-for-process trade-off method rather than an EUV substitute, and the piece’s own metadata pairs this with the piece’s own diagnostic that China still relies on DUV lithography. The same chip-manufacturing surface this series’ September 30 briefing documented at the Huawei-Ascend-Tribe-plus-openPangu-2.0-Training-plus-openPangu-2.0-RL layer is now being operationalized at the LogicFolding-plus-area-for-process-trading-plus-not-EUV-substitute layer simultaneously with the Huawei-has-essentially-broken-free-plus-China-still-relies-on-DUV layer — a two-layer-simultaneous Huawei-chip-architecture-vs-national-industry-structural-fact institutional-form observation.

The OpenHarmony-90-million-devices-plus-11-million-developers-plus-75-percent-foldable-phone-market-share layer. Per the piece: OpenHarmony device count 90 million+ (expected to break 100 million by year-end); OpenHarmony application-and-service ecosystem 450,000+; OpenHarmony global-registered-developer count 11 million+; Chinese-foldable-phone market share 75%; Mate XT 2 has shipped approximately 100,000 units in the first 15 days after launch. The institutional reading is that the OpenHarmony-90-million-devices-plus-11-million-developers-plus-75-percent-foldable-phone-market-share layer is being operationalized at the vendor-official-plus-industry-structural-fact-plus-OpenHarmony-device-count-and-developer-count-and-ecosystem-count layer — the same OpenHarmony ecosystem-surface this series’ September 8 briefing documented at the Wang Chenglu dialogue’s 14-billion-device-base-layer-plus-90-to-70-percent-contribution-share-decline layer is now being operationalized at the 90-million-devices-plus-11-million-developers-plus-75-percent-foldable-phone-market-share layer. The OpenHarmony device count reported at the Wang Chenglu dialogue (14 billion device installs, 3 billion of which are Huawei phones) versus the 90 million-device figure reported at the Yu roundtable is a significant discrepancy between two Huawei-official-surface reports — and the institutional reading is that the two reported OpenHarmony device counts at the two institutional surfaces are structurally distinct institutional-form objects: the Wang Chenglu 14 billion device count is characterized as an OpenHarmony-ecosystem device count including non-Huawei-device installs, while the Yu 90 million device count is characterized as an OpenHarmony-device count at the Huawei Consumer BG level — a two-layer-simultaneous OpenHarmony-device-count-institutional-form observation.

The 自主可控新鲜事 selective-headline-framing layer. Per the WeChat input metadata: the piece’s own headline is characterized as “选择性强调’基本摆脱’（华为公司层面）弱化了’仍依靠DUV’（中国产业层面）” (selectively emphasizes “essentially broken free” (Huawei-company layer) and weakens “still relies on DUV” (Chinese-industry layer)). Per the piece’s own metadata characterization: “这是行政开源叙事的经典传播机制——市场化媒体+行政叙事；市场化媒体不需要行政指令，只需要选择性传播符合自主可控叙事的标题” (This is the classic propagation mechanism of the administrative-open-source narrative — market-media-plus-administrative-narrative; market-media does not need administrative instructions; it only needs to selectively propagate headlines that fit the autonomous-controllable narrative). The institutional reading is that the 自主可控新鲜事 selective-headline-framing layer is being operationalized at the autonomous-controllable-narrative-plus-market-media-selective-transmission institutional-form layer — a first-documented instance in this series of the autonomous-controllable-narrative’s market-media-plus-administrative-narrative selective-transmission mechanism being operationalized at the individual-vendor-vs-national-industry dichotomy layer. The market-media-plus-administrative-narrative selective-transmission mechanism is a structurally distinct institutional-form object from the state-directed-industrial-policy-instrument layer this series’ September 14 briefing documented.

The institutional-object pairing with this series’ September 8 briefing’s OpenHarmony-14-billion-device-base-layer-plus-90-to-70-percent-contribution-share-decline layer. This series’ September 8 briefing documented the OpenHarmony base layer at the 14-billion-device-plus-90-to-70-percent-contribution-share-decline-plus-Hubble-20-percent-stake-plus-2028-ecosystem-platform-target layer. This cycle’s briefing documents the same OpenHarmony ecosystem surface at the 90-million-devices-plus-11-million-developers-plus-450000-applications-plus-75-percent-foldable-phone-market-share layer — a two-layer-simultaneous OpenHarmony-device-count-and-ecosystem-count institutional-form observation.

The institutional-object pairing with this series’ October 1 briefing’s Xinchuang-tail-mid-tier-vendor-exit-and-1200-SOE-digital-technology-subsidiary-collapse layer. The Yu statement’s own characterization that “华为作为高端厂商自有工艺厂商是分化期中唯一受益者” (Huawei as a high-end vendor with its own process capability is the sole beneficiary in the stratification period) — paired with this cycle’s briefing’s 国产信创外传 piece’s own diagnostic that “华为在信创分化期中是唯一受益的厂商” — is a structurally new first-documented instance in this series of the same vendor (Huawei) being positioned at the head-of-Xinchuang-stratification layer at two institutionally distinct surfaces (the vendor-official roundtable surface and the state-informatization-industry-outlet surface) simultaneously with the same Xinchuang-institutional-object’s tail-mid-tier-vendor-exit-and-1200-SOE-digital-technology-subsidiary-collapse layer.

The institutional-economics question this cycle’s briefing documents but does not answer. Whether the autonomous-controllable-narrative-plus-market-media-selective-transmission institutional-form object this cycle’s briefing documents at the 自主可控新鲜事-selective-headline-framing layer converts the same vendor-official-plus-industry-structural-fact surface this cycle’s briefing documents at the individual-vendor-vs-national-industry dichotomy layer into a specific operational outcome at the autonomous-controllable-narrative-plus-market-media-selective-transmission layer, and whether the same vendor (Huawei) being positioned at the head-of-Xinchuang-stratification layer at two institutionally distinct surfaces simultaneously with the same Xinchuang-institutional-object’s tail-mid-tier-vendor-exit-and-1200-SOE-digital-technology-subsidiary-collapse layer converts the Xinchuang-institutional-object’s head-and-tail configuration into a specific operational outcome at the same institutional-object layer.

Institutional significance: Richard Yu’s September 29, 2026 international-media roundtable plus the 自主可控新鲜事’s selective-headline-framing is the first-documented instance in this series of the autonomous-controllable-narrative-plus-market-media-selective-transmission institutional object being operationalized at the individual-vendor-vs-national-industry-dichotomy layer simultaneously with this series’ October 8 briefing’s Huawei-Ascend-CANN-external-contributor-majority layer, this series’ September 30 briefing’s Huawei-Ascend-Tribe-openPangu-2.0-Training-plus-openPangu-2.0-RL layer, and this series’ September 18 briefing’s HUAWEI CONNECT CANN-community-open-source-development layer — a three-surface-simultaneous Huawei-narrative-plus-vendor-institutional-form-plus-autonomous-controllable-narrative-structural-fact institutional-form observation.

Sources:

🔍 Commentary

Three institutionally dense placements on the Chinese open-source institutional-form surface this cycle’s briefing documents, each of which sits at an institutional layer this series’ prior briefings have not documented at before, and all three placements sit at institutional layers that pair structurally with this series’ prior briefings’ documentation of the same institutional object:

  1. OpenHarmony’s 《生成式AI工具使用与开源贡献策略》 at the AI-governance-rule-execution-plus-four-dimensions-of-community-governance (manual-review, full-test-suite, transparent-disclosure, tiered-verification) plus two-person-cross-verification-for-high-severity-vulnerabilities-plus-prohibition-on-AI-inference-directly-determining-risk-grade layer (2026-10-08). This is a first-documented instance in this series of a large Chinese operating-system open-source community’s official community channel being operationalized at the AI-governance-rule-execution layer simultaneously with this series’ September 8 briefing’s OpenHarmony-14-billion-device-base-layer-plus-90-to-70-percent-contribution-share-decline layer and this series’ September 14 briefing’s state-directed-open-source-project-selection-and-community-first-release layer.

  2. 国产信创外传’s Xinchuang-blind-replacement-slowdown-plus-business-first-value-based-rollout-plus-Huawei-as-sole-beneficiary at the state-informatization-industry-outlet-plus-value-based-rollout-reframe-plus-three-principles (tiered-replacement, assess-before-modify, gray-rollout-dual-track-verification) plus three-institutional-risks-being-avoided (business-interruption, hidden-adaptation, long-term-O&M) plus operational-objective-reframe layer. This is a first-documented instance in this series of the Xinchuang-blind-replacement-slowdown institutional object being operationalized at the state-informatization-industry-outlet-plus-value-based-rollout-plus-Huawei-as-sole-beneficiary institutional-form layer simultaneously with this series’ October 1 briefing’s Xinchuang-tail-mid-tier-vendor-exit-and-1200-SOE-digital-technology-subsidiary-collapse layer.

  3. Richard Yu’s September 29, 2026 international-media roundtable plus the 自主可控新鲜事’s selective-headline-framing at the Huawei-has-essentially-broken-free-plus-China-still-relies-on-DUV-plus-LogicFolding-as-area-for-process-trading-method-plus-not-EUV-substitute-plus-autonomous-controllable-narrative-selective-transmission institutional-form layer. This is a first-documented instance in this series of the autonomous-controllable-narrative-plus-market-media-selective-transmission institutional object being operationalized at the individual-vendor-vs-national-industry-dichotomy layer simultaneously with this series’ October 8 briefing’s Huawei-Ascend-CANN-external-contributor-majority layer, this series’ September 30 briefing’s Huawei-Ascend-Tribe-openPangu-2.0-Training-plus-openPangu-2.0-RL layer, and this series’ September 18 briefing’s HUAWEI CONNECT CANN-community-open-source-development layer.

One structural pattern across the three placements: the same unsolved institutional object — the boundary between Chinese vendor-official-plus-community-official-and-community-governance-plus-narrative-selective-transmission surfaces and the FLOSS-commons institutional identity this series’ prior briefings have been documenting across institutional borders — is being operationalized at three institutionally distinct layers simultaneously in this cycle’s briefing. The OpenHarmony community’s AI-tool-use rulebook operationalizes the boundary at the large-Chinese-OS-community-official-channel-plus-community-governance-instrument layer; the 国产信创外传 piece operationalizes the boundary at the state-informatization-industry-outlet-plus-value-based-rollout-reframe-plus-Huawei-as-sole-beneficiary layer; and the 自主可控新鲜事 piece operationalizes the boundary at the autonomous-controllable-narrative-plus-market-media-selective-transmission-plus-individual-vendor-vs-national-industry-dichotomy layer. A three-surface-simultaneous Chinese-institutional-form boundary observation that this series’ prior briefings have not documented at before.

One structural risk across the three placements: the same boundary institutional object this cycle’s briefing documents on three institutionally distinct layers may convert the Chinese-institutional-form boundary from an institutional-form object into a specific operational outcome object — the OpenHarmony community’s AI-governance-rule-execution layer into a community-governance-instrument-object rather than a platform-vendor-instrument-object; the Xinchuang-institutional-object into a specific head-and-tail stratification outcome rather than a state-directed-replacement-rate-instrument; and the autonomous-controllable-narrative-selective-transmission into a specific propagation-mechanism-outcome rather than a market-media-independent-reporting-object. The three-surface-simultaneous institutional-form-translation pattern is the same pattern this series’ October 8 briefing documented at the NDA-embodied-AI-standards-plus-agentic-AI-framework layer and the DeepSeek-plus-Huawei-Ascend-sovereign-hardware-stack layer.

  • The OpenHarmony AI-tool-use rulebook placement pairs institutionally with this series’ September 8 briefing’s OpenHarmony-14-billion-device-base-layer-plus-90-to-70-percent-contribution-share-decline layer and this series’ September 14 briefing’s state-directed-open-source-project-selection-and-community-first-release layer.
  • The 国产信创外传 Xinchuang-blind-replacement-slowdown placement pairs institutionally with this series’ October 1 briefing’s Xinchuang-tail-mid-tier-vendor-exit-and-1200-SOE-digital-technology-subsidiary-collapse layer.
  • The 自主可控新鲜事 selective-headline-framing placement pairs institutionally with this series’ October 8 briefing’s Huawei-Ascend-CANN-external-contributor-majority layer, this series’ September 30 briefing’s Huawei-Ascend-Tribe-openPangu-2.0-Training-plus-openPangu-2.0-RL layer, and this series’ September 18 briefing’s HUAWEI CONNECT CANN-community-open-source-development layer.

The institutional-economics question — which this cycle’s briefing documents but does not answer — is whether the Chinese-institutional-form boundary object this cycle’s briefing documents on three institutionally distinct layers simultaneously — the OpenHarmony AI-tool-use rulebook layer, the Xinchuang-blind-replacement-slowdown-and-value-based-rollout-reframe layer, and the autonomous-controllable-narrative-selective-transmission layer — can hold the same boundary object at three institutionally distinct layers without one converting the other, and whether the same three-surface-simultaneous institutional-form-translation pattern this cycle’s briefing documents signals a broader Chinese-institutional-form-boundary object being operationalized at the community-governance-instrument-plus-state-informatization-industry-outlet-plus-market-media-selective-transmission layer.

One perspective, not a verdict.

The three placements this cycle’s briefing documents — OpenHarmony’s AI-tool-use rulebook, the 国产信创外传’s Xinchuang-blind-replacement-slowdown, and the 自主可控新鲜事’s selective-headline-framing — are best read as observations of institutional movement in progress, not as verdicts on institutional direction.

Editorial note on perspective: This briefing presents one institutional-economics reading of Chinese open-source developments, not a verdict. The “institutions” in this story — the OpenHarmony community’s rulebook, the 国产信创外传’s operational reframe, and the 自主可控新鲜事’s selective-headline-framing — are treated as objects of observation, not targets of critique. The Great Divergence 2.0 framework (FLOSS vs. State-Chartered Codebase vs. Intranet Shared Source vs. Cyber-Eestate) and the Williamson L1→L4 institutional-economics reading (L1 social embedding → L2 institutional environment → L3 governance mechanisms → L4 resource allocation) are lenses, not universal answers. One perspective, not a verdict.

Deduplication note: The three stories this cycle’s briefing documents — OpenHarmony’s 《生成式AI工具使用与开源贡献策略》 (2026-10-08), the 国产信创外传’s Xinchuang-blind-replacement-slowdown-and-value-based-rollout-reframe-plus-Huawei-as-sole-beneficiary (2026-10-08), and the 自主可控新鲜事’s re-publication of Richard Yu’s September 29, 2026 international-media roundtable (2026-10-08) — were not covered in any of the prior three briefings (October 6, October 7, October 8). The October 6 briefing covered the State Council Decree 841 Exit and Entry Administration Regulation plus Bloomberg’s September 28 family-extension reporting, President Trump’s August 26 executive order plus the U.S.-Senate-bill-barring-federal-agencies-from-buying-Chinese-optical-transceivers reporting, the Trump-August-26-power-system-executive-order layer, and the ByteDance-and-Alibaba-AI-companion-withdrawal-plus-new-rules-loom layer. The October 7 briefing covered CNBC’s DeepSeek-15-billion-dollar-funding-round-consideration documentation, Reuters’s DeepSeek-12-billion-funding-round documentation, Euronews’s Moonshot-50-billion-valuation-Hong-Kong-IPO-eyes documentation, the South China Morning Post HKIC-BAAI-global-AI-talent-partnership documentation, and ChinaTalk’s China-AI-safety-money-problem documentation. The October 8 briefing covered The Diplomat’s October 2026 essay on China’s AI-governance push and the race to shape global AI rules, Forkast’s The Deployment State essay on China’s state-directed AI-agent industrial-policy apparatus, and Crypto Briefing’s DeepSeek-plus-Huawei-Ascend-950DT-plus-Ulanqab-160000-chip-1-gigawatt-plus-Ascend-960DT-launch-acceleration documentation. None of the prior three briefings covered the OpenHarmony community’s formal AI-tool-use rulebook, the 国产信创外传’s Xinchuang-blind-replacement-slowdown-and-value-based-rollout-reframe, or the 自主可控新鲜事’s selective-headline-framing-plus-individual-vendor-vs-national-industry-dichotomy layer.

WeChat input note: The raw/2026-10-08.md file was found in the china-daily-english-input repository. The WeChat section of this cycle’s briefing documents three WeChat-channel articles: (1) 开放原子开源基金会 / 开源鸿蒙社区’s OpenHarmony AI-tool-use rulebook; (2) 国产信创外传’s Xinchuang-blind-replacement-slowdown-and-value-based-rollout-reframe; (3) 自主可控新鲜事’s re-publication of Richard Yu’s September 29, 2026 international-media roundtable.