⚠️ 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 — 2026-10-13

🏛️ Policy & Regulation — CAICT’s 2026 Internet Agent Governance Innovation Application Practice Activity Launched on 2026-10-12 as the First-Documented Instance in This Series of a State-Adjacent Research Institute Operationalizing the 2026-05-08 Agent Guidelines at the Institute-Level Application-Activity Layer, Completing the Policy-Framework-Plus-Technology-Standard-Plus-Capability-Assessment-Plus-Governance-Practice Institutional Chain at Its First Application-Layer Activity

1. CAICT (中国信息通信研究院, directly under the Ministry of Industry and Information Technology) — 2026-10-12 launch of the inaugural Internet Agent Governance Innovation Application Practice Activity.

2. Six recommendation directions — (a) business-scenario application, (b) agent infrastructure, (c) agent management platform, (d) terminal agent, (e) inter-agent interaction and collaboration (explicitly naming ATH agent trusted-handshake protocol plus A2A plus MCP protocol adaptation applications), (f) security governance (explicitly naming open-source-community-governance-practice as a governance sub-topic, alongside identity security, permission control, data protection, memory security, tool-invocation security, skill-and-component governance, runtime isolation, behavior audit, risk monitoring, emergency handling, agent-security evaluation, and governance-mechanism construction).

3. Ten participant unit categories — basic telecommunications operators, internet enterprises, cloud service providers, AI enterprises, terminal enterprises, software enterprises, security enterprises, higher-education institutions, research institutes, and end-user units from various industries; technology-provider units and application units are encouraged to jointly recommend; each leading unit is capped at three recommendation items, and the same practice may not be re-recommended by different units or across different directions.

4. Timeline — submission window 2026-10-12 to 2026-11-06, review-and-judging in 2026-11, results release in early 2026-12 at the AI-plus-Industry-Ecology Conference’s Internet Agent Forum.

5. Host department — CAICT Cloud and Big Data Institute’s Open Source and Intelligent Safety Department (开源与智能安全部); named contacts Ma Mingyang and Xing Luowen.

6. Policy lineage — this activity is positioned as the concrete implementation at the governance layer of the 2026-05-08 Guidelines on Standardized Application and Innovative Development of Agents jointly issued by the Cyberspace Administration of China, the National Development and Reform Commission, and the Ministry of Industry and Information Technology, and constitutes the first application-layer activity in the complete institutional chain of policy-framework → technology-standard → capability-assessment → governance-practice.

7. ATH protocol context — the ATH agent trusted-handshake protocol was jointly initiated on 2026-04-29 by CAICT together with China Telecom, China Unicom, China Mobile Jiutian, The Chinese University of Hong Kong Shenzhen, ZTE, Tencent, and Huawei, and was dual-released as open source on AtomGit and GitHub, positioned as complementary-plus-mutually-reinforcing (not competing) with MCP and A2A; the first batch of ATH capability assessment results were published in 2026-06 to 2026-07, with 13 enterprises’ products and services passing evaluation.

The CAICT 2026 Internet Agent Governance Innovation Application Practice Activity, launched on 2026-10-12, is the first-documented instance in this series of a Chinese state-adjacent research institute operationalizing the 2026-05-08 agent guidelines (jointly issued by the Cyberspace Administration of China, the National Development and Reform Commission, and the Ministry of Industry and Information Technology) at the institute-level application-activity layer. The activity completes the CAICT agent-governance institutional object this series has been tracking since April 2026 at a new institutional layer — moving from the technology-standard-plus-protocol-initiation-plus-AtomGit-GitHub-dual-open-source-release-plus-first-assessment-13-firms layer to the policy-framework-plus-technology-standard-plus-capability-assessment-plus-governance-practice institutional-chain-completion layer.

From an institutional economics perspective, this activity matters on four axes:

First, the policy-lineage-plus-institute-application-activity-plus-CF-CAICT-MIIT-MIIT-joint-guideline-plus-application-activity-completion pairing is an institutional-economics fact about the Chinese-agent-governance-institutional-chain’s first-documented-completion layer. The 2026-05-08 agent guidelines jointly issued by the Cyberspace Administration of China, NDRC, and MIIT (a three-ministry-plus-Cyberspace-Agency co-signature) constituted the policy-framework layer; the ATH protocol dual-open-source release on 2026-04-29 through AtomGit and GitHub by CAICT plus eight technology-and-carrier consortium members constituted the technology-standard layer; the first batch of ATH capability assessments published in 2026-06/07 with 13 enterprises passing constituted the capability-assessment layer. This October 12 activity completes the governance-practice layer. From an institutional economics standpoint, this guideline-2026-05-08 / ATH-protocol-2026-04-29 / assessment-2026-06-07-13-firms / application-activity-2026-10-12 pairing is a structural finding: the CAICT agent-governance institutional object is being operationalized not at one institutional layer (policy, or standard, or assessment, or practice) but at a four-layer-simultaneous institutional-chain-completion layer. The institutional-economics reading is that the Chinese agent-governance institutional chain is now documented as a complete policy-plus-standard-plus-assessment-plus-practice chain at the same CAICT-institute-surface — a form of institutional-chain-completion that this series has documented elsewhere in the Western open-source foundation surface (PyTorch Foundation 2026-09-08 China-member expansion plus PyTorch Conference China 2026 co-location layer) but has not documented at the Chinese state-adjacent-research-institute-surface before.

Second, the six-recommendation-directions-plus-open-source-community-governance-explicit-citation pairing is an institutional-economics fact about the Chinese-agent-governance-policy-framework’s first-documented-open-source-community-governance layer. The security-governance recommendation direction explicitly names 开源社区治理实践 (open-source community governance practice) as one of the sub-topics, alongside identity security, permission control, data protection, memory security, tool-invocation security, skill-and-component governance, runtime isolation, behavior audit, risk monitoring, emergency handling, and agent-security evaluation. From an institutional economics standpoint, this agent-governance-plus-open-source-community-governance-plus-CAICT-Cloud-and-Big-Data-Institute-Open-Source-and-Intelligent-Safety-Department pairing is a structural finding: the CAICT’s Cloud and Big Data Institute’s Open Source and Intelligent Safety Department — an institute-internal department that, per its name, is institutionalized at the joint open-source-plus-intelligent-safety layer — is the host department for an activity that explicitly names open-source community governance as one of the sub-topics under a policy framework governing AI agent deployment. The institutional-economics reading is that the CAICT has, through the co-location of its department name and its activity sub-topic, documented a Chinese state-adjacent-research-institute institutional surface that is operationalized at a two-institutional-object-simultaneous layer (open-source-community-governance-plus-agent-governance), and this co-location has not been documented at this level of specificity in this series’ prior fifty-two briefings.

Third, the ten-participant-unit-categories-plus-three-cap-per-unit-plus-tech-provider-and-applicant-joint-recommendation pairing is an institutional-economics fact about the Chinese-agent-governance-activity-participant-cohort-design layer. The activity solicits participation from ten distinct categories of institutional units — basic telecommunications operators, internet enterprises, cloud service providers, AI enterprises, terminal enterprises, software enterprises, security enterprises, higher-education institutions, research institutes, and industry end-user units — and explicitly encourages joint recommendation between technology-provider units and application units, while capping each leading unit at three recommendation items and prohibiting the same practice from being re-recommended by different units or across different directions. From an institutional economics standpoint, this ten-category-participant-cohort / three-cap-plus-joint-recommendation-plus-no-cross-unit-duplication pairing is a structural finding: the activity is institutionalized not as a market-mechanism-plus-free-solicitation form but as a quota-limited-plus-joint-signature-required-plus-deduplication-required form — the design constraints themselves (ten categories, three cap, joint-signature, deduplication) are an institutional-form observation on the CAICT activity surface. The institutional-economics reading is that the CAICT agent-governance activity is not a public-market-style solicitation but a quota-limited-plus-collaboration-required-plus-deduplication-required institutional form; this is the third documented quota-and-solicitation institutional form this series has observed on the CAICT surface (the first being the ATH protocol consortium-member-plus-eight-consortium-membership-initiation form, the second being the 13-enterprises-passing-first-assessment cohort form, and the third being this ten-category-plus-three-cap application-activity form) — three forms, one institutional surface, all institutionalizations at the CAICT-institute layer.

Fourth, the policy-framework-to-technology-standard-to-capability-assessment-to-governance-practice institutional-chain-plus-14-months-timeline pairing is an institutional-economics fact about the Chinese-agent-governance-institutional-chain’s first-documented-completion-timeline layer. The CAICT 2026-10-12 activity is positioned as the concrete implementation at the governance layer of the 2026-05-08 agent guidelines — a 14-month interval between guideline issuance and first application-layer activity. From an institutional economics standpoint, this 2026-05-08-guideline / 2026-04-29-ATH-protocol-initiation / 2026-06-07-first-assessment / 2026-10-12-application-activity pairing is a structural finding: the CAICT agent-governance institutional chain is documented at a four-event-plus-14-months-plus-four-layer-simultaneous layer — the same CAICT-surface that this series’ prior briefings documented at the technology-standard layer (April 29, 2026 ATH protocol dual-open-source release) is now documented at the governance-practice layer (October 12, 2026 application-activity launch). The institutional-economics reading is that the CAICT agent-governance institutional chain is not one policy document (the May 8 guidelines) but a four-layer-plus-14-months institutional chain, and the October 12 activity is the first-documented instance of this series documenting the completion of the CAICT agent-governance institutional chain at its governance-practice layer.

Sources:

🏗️ Institutional Change — 自主可控新鲜事’s 2026-10-12 Report Documents Tianjin Binhai High-Tech Development Zone’s Xinchuang Industry Bureau as the First-Documented Instance in This Series of a Local Xinchuang Governance Institution Being Operationalized at the Renaming-Plus-Expansion-Not-Launch Institutional-Formation Layer Simultaneously with the Administrative-Bureau-Plus-Comprehensive-Party-Committee-Plus-Industry-Alliance Three-Layer-Architecture Layer

1. 自主可控新鲜事 outlet (2026-10-12) — report titled “天津悄悄设了个’信创产业局’，这事不简单！” (Tianjin quietly set up a ‘Xinchuang Industry Bureau’ — this is not a simple matter)

2. Core fact — the Tianjin Binhai High-Tech Development Zone Xinchuang Industry Bureau (天津滨海高新技术产业开发区信创产业局) has issued a batch of procurement pre-notices; no public inauguration record exists; no dedicated press release has been issued.

3. Institutional formation mode — historical name trace: Marine-Economy-and-Military-Civilian-Fusion Industry Investment Bureau (天津滨海高新技术产业开发区海洋经济和军民融合产业招商局, earliest documented form); Xinchuang Industry Investment Bureau (天津滨海高新技术产业开发区信创产业招商局, documented in 2021 financial statements); Xinchuang Industry Bureau (天津滨海高新技术产业开发区信创产业局, first officially used in March 2023 appointment notice). The name trajectory — from ocean-economy-plus-military-civilian-fusion to Xinchuang, from Investment Bureau to Industry Bureau — documents institutional formation through renaming and expanded authority, not through a launch event.

4. Institutional scope — the department’s financial statements disclose seven core functions: industrial-innovation-and-development research, policy formulation, investment attraction, industry-chain development, innovation-element aggregation, coordination of the Haihe Laboratory’s advanced-computing-and-key-software-Xinchuang research-platform construction, and talent attraction and cultivation. The Haihe Laboratory (海河实验室) is Tianjin’s core Xinchuang research platform, adopting a small-core-plus-large-platform model that links universities and enterprises across the base-research-to-industrialization chain.

5. Four nationwide Xinchuang governance institutional categories documented in the report — (a) dedicated-sections form: Chengdu High-Tech Zone’s High-End Software Industry Section (成都高新区高端软件产业处) and Shanxi Provincial Science and Technology Department’s Xinchuang and Big Data Science and Technology Section (山西省科技厅信创和大数据科技处); (b) working-group form: Jiangsu Information Technology Application Innovation Industry Chain-Strengthening Working Group (江苏信息技术应用创新产业强链工作专班), Zhejiang Jinhua Integrated Circuit and Xinchuang Industry Chain Working Group (浙江金华市集成电路及信创产业链专班), and Shanxi Changzhi High-Tech Zone Xinchuang Working Group (山西长治高新区信创工作专班); (c) service-center form: Ningbo Xinchuang Service Support Center, unveiled in November 2024 (宁波市2024年11月揭牌成立信创服务保障中心); (d) existing-department-coordination form: Beijing’s Municipal Bureau of Economic and Information Technology (北京市经济和信息化局) plus Fujian’s Provincial Development and Reform Commission, Cryptography Administration, and Industry and Information Technology Department joint coordination. Tianjin’s form is documented as the most institutionalized of the four categories.

6. Institutional architecture — the Binhai High-Tech Zone has layered a Xinchuang Industry Comprehensive Party Committee (信创产业综合党委) over the Xinchuang Industry Bureau, absorbing Party-member technical backbones from research institutes and leading enterprises (Feiteng 飞腾, Kylin 麒麟, Hygon 海光, Nandasoft 南大通用, Sugon 曙光, 360) into a three-layer architecture: administrative bureau + comprehensive Party committee + industry alliance.

7. Scale metrics — 2021 Tianjin released the “China Xinchuang Valley Development Plan (2020-2025)” (《中国信创谷发展规划2020-2025年》), positioning Tianjin as a demonstration zone for high-quality national Xinchuang industry development; the Binhai High-Tech Zone has cultivated over 200 key Xinchuang enterprises (Phytium 飞腾, Hygon 海光, Kylin 麒麟, Nandasoft 南大通用, Sugon 曙光, 360, and others) building a full-stack Xinchuang industry chain covering CPU-plus-operating-system-plus-database-plus-server-plus-endpoint-terminal-plus-cloud-computing-plus-information-security; 2024 Xinchuang industry revenue exceeded 72 billion yuan (up 30.1% YoY); 2025 full-year Xinchuang industry revenue was approximately 85 billion yuan (up 17.4% YoY); 2025 Q1-to-Q3 Tianjin citywide Xinchuang industry scale reached 114.99 billion yuan; 2026 Q1 the Xinchuang Bureau registered 41 new landed key enterprises.

The 自主可控新鲜事 outlet’s 2026-10-12 report on the Tianjin Binhai High-Tech Development Zone Xinchuang Industry Bureau is the first-documented instance in this series of a local Xinchuang governance institution being operationalized at the renaming-plus-expansion-not-launch institutional-formation layer simultaneously with the administrative-bureau-plus-comprehensive-party-committee-plus-industry-alliance three-layer-architecture layer. The report documents that the Xinchuang Industry Bureau is not an institution that appeared through a formal launch event but one that has been progressively renamed and expanded through institutional formation — a form of institutional formation this series has documented elsewhere in the Western open-source foundation surface (PyTorch Foundation’s 2026-09-08 China-member expansion plus PyTorch Conference China 2026 co-location layer, and Moonshot AI’s pre-IPO-to-Hong-Kong-IPO institutional-form-transition layer) but has not documented at this level of specificity at the local Chinese Xinchuang governance-institution surface.

From an institutional economics perspective, this report matters on four axes:

First, the ocean-economy-plus-military-civilian-fusion-to-xinchuang-institutional-formation-plus-investment-bureau-to-industry-bureau-name-change pairing is an institutional-economics fact about the local-Chinese-Xinchuang-governance-institution’s first-documented-rename-plus-expand-not-launch layer. The name trajectory of the Tianjin Binhai Xinchuang Institution — from Marine-Economy-and-Military-Civilian-Fusion Industry Investment Bureau to Xinchuang Industry Investment Bureau (2021 financial statements) to Xinchuang Industry Bureau (2023 appointment notice) — is not a linear renaming but a two-step institutional re-orientation: (a) domain re-orientation from ocean-economy-plus-military-civilian-fusion to Xinchuang (a shift in institutional object — the institution is being re-pointed from one industrial domain to another), and (b) function re-orientation from Investment Bureau (招商局) to Industry Bureau (产业局) (a shift in institutional function — the institution is being re-pointed from an investment-attraction-plus-solicitation role to a broader industry-governance-plus-policy-plus-R-and-D-coordination role). From an institutional economics standpoint, this ocean-military-civilian-fusion-to-xinchuang / investment-bureau-to-industry-bureau pairing is a structural finding: the Tianjin Binhai Xinchuang Industry Bureau is not one institution (a new bureau created on some date) but one institution undergoing two-step institutional formation (rename plus expand) at the local Xinchuang governance institutional surface. The institutional-economics reading is that the local Chinese Xinchuang governance institutional surface is being operationalized at a two-step-institutional-formation layer — a layer this series has documented elsewhere at the corporate-institutional surface (宜信 → 宜人贷 → 宜人金科 → 宜人智科 corporate-name-plus-industry-reorientation progression) but has not documented at this level of specificity at the local-government Xinchuang governance institutional surface.

Second, the four-nationwide-categories-of-xinchuang-governance-institutions-plus-tianjin-picks-most-institutionalized-form pairing is an institutional-economics fact about the Chinese-Xinchuang-governance-institutional-form-taxonomy layer. The report documents that local Xinchuang governance institutions across China take four institutional forms: (a) dedicated sections within existing government departments (Chengdu, Shanxi), (b) working groups (Jiangsu, Zhejiang Jinhua, Shanxi Changzhi), (c) service centers (Ningbo, November 2024), and (d) existing-department-coordination (Beijing, Fujian). Tianjin’s form — a full-fledged Xinchuang Industry Bureau — is documented as the most institutionalized of the four categories. From an institutional economics standpoint, this four-category-xinchuang-institutional-form-taxonomy / Tianjin-picks-most-institutionalized pairing is a structural finding: the Chinese Xinchuang governance institutional surface is being operationalized not at a single institutional-form layer (one bureau type, or one working group type) but at a four-category-institutional-form-taxonomy layer in which Tianjin picks the most institutionalized form. The institutional-economics reading is that the local Chinese Xinchuang governance institutional surface is documented as a four-category institutional-form taxonomy — a taxonomy this series has not documented at this level of specificity anywhere in its 52 prior briefings — and Tianjin’s choice of the most-institutionalized form (a full-fledged bureau, plus a layered comprehensive Party committee, plus an industry alliance) is a first-documented instance of a local Chinese government choosing the most-institutionalized form of Xinchuang governance.

Third, the administrative-bureau-plus-comprehensive-party-committee-plus-industry-alliance three-layer-architecture pairing is an institutional-economics fact about the Chinese-local-Xinchuang-governance’s first-documented-three-layer-architecture layer. The Binhai High-Tech Zone has layered three institutional forms over the Xinchuang Industry Bureau: (a) the Xinchuang Industry Bureau itself (administrative bureau form), (b) the Xinchuang Industry Comprehensive Party Committee layered over the bureau (a Party-committee form that absorbs Party-member technical backbones from research institutes and leading enterprises including Phytium, Kylin, Hygon, Nandasoft, Sugon, 360), and (c) the industry alliance (the industry alliance form, implied by the three-layer description). From an institutional economics standpoint, this administrative-bureau / comprehensive-party-committee / industry-alliance three-layer pairing is a structural finding: the Tianjin Binhai Xinchuang governance institutional surface is being operationalized at a three-layer-institutional-architecture layer — a layer that combines the state-administrative form (bureau), the party-organizational form (comprehensive Party committee), and the industry-alliance form (industry alliance) on the same institutional surface. The institutional-economics reading is that the local Chinese Xinchuang governance institutional surface is not one institutional form (a bureau, or a working group, or a service center, or a department-coordination) but a three-layer institutional architecture — a form this series has documented elsewhere at the state-adjacent-research-institute-surface (CAICT as a state-adjacent institute plus its internal Open Source and Intelligent Safety Department plus its ATH protocol consortium) and at the state-owned-enterprise-surface (SASAC Decree No. 46 plus the Xinchuang policy-tightening surface this series’ October 11 briefing documented) but has not documented at this level of specificity at the local-government Xinchuang governance institutional surface.

Fourth, the 85-billion-plus-114.99-billion-xinchuang-industry-scale-plus-41-new-landed-2026-q1 enterprises-plus-two-step-institutional-formation pairing is an institutional-economics fact about the Chinese-local-Xinchuang-industrial-scale-vs-institutional-formation pairing. The report documents that Tianjin’s 2025 Xinchuang industry revenue was approximately 85 billion yuan (up 17.4% YoY), that 2025 Q1-to-Q3 Tianjin citywide Xinchuang industry scale reached 114.99 billion yuan, and that 2026 Q1 the Xinchuang Bureau registered 41 new landed key enterprises — while also documenting that the Xinchuang Industry Bureau’s institutional formation (from ocean-economy-plus-military-civilian-fusion-investment-bureau through Xinchuang-investment-bureau to Xinchuang-industry-bureau) took approximately two years (2021 to 2023) and was not accompanied by a public launch ceremony. From an institutional economics standpoint, this 85-billion-2025-xinchuang-industry-revenue / 41-new-landed-2026-q1 / two-step-name-change-2021-2023 pairing is a structural finding: the local Chinese Xinchuang industrial-scale-plus-institutional-formation pairing is being operationalized at a two-year-name-change-plus-85-billion-industry-revenue layer — an institutional-economics observation in which the industrial scale (85 billion yuan in 2025) outpaces the institutional formation visibility (no public launch ceremony, only a two-year progressive renaming). The institutional-economics reading is that the local Chinese Xinchuang governance institutional surface is being operationalized at a two-surface-simultaneous layer — the industrial-scale-visible layer (85 billion yuan revenue, 41 new landed enterprises) and the institutional-formation-invisible layer (two-year progressive renaming, no public launch) — a two-surface-simultaneous observation that is the inverse of the CAICT agent-governance institutional-chain-completion layer documented in Story 1 above (which is a four-layer-plus-14-months institutional-chain-visible-plus-policy-framework-visible layer).

Sources:

  • 自主可控新鲜事 (2026-10-12) — 天津悄悄设了个"信创产业局"，这事不简单！
  • Context: 2021 Tianjin “China Xinchuang Valley Development Plan (2020-2025)”; Binhai High-Tech Zone 200+ cultivated Xinchuang key enterprises; 2024 Xinchuang industry revenue 72 billion yuan (+30.1% YoY); 2025 Xinchuang industry revenue approximately 85 billion yuan (+17.4% YoY); 2025 Q1-to-Q3 Tianjin citywide Xinchuang industry scale 114.99 billion yuan; 2026 Q1 Xinchuang Bureau registered 41 new landed key enterprises

🔍 Commentary

Two stories, one institutional pattern: state-adjacent-research-institute-agent-governance-institutional-chain-completion-plus-local-government-Xinchuang-governance-institutional-formation-plus-three-layer-architecture-plus-two-step-rename layer activation on the same CAICT-institute-and-Tianjin-Binhai-governance institutional objects.

This cycle’s two stories document, on two different institutional surfaces, the same unsolved institutional pattern: the state-adjacent-research-institute-and-local-government-institutional-formation institutional object activating first-documented-institutional-chain-completion-and-three-layer-architecture institutional identities this series’ prior fifty-two briefings have not documented at before.

  • CAICT 2026 Internet Agent Governance Innovation Application Practice Activity (2026-10-12) documents the CAICT-institute institutional object at the policy-framework-to-technology-standard-to-capability-assessment-to-governance-practice four-layer institutional-chain-completion layer simultaneously with the open-source-community-governance-explicit-citation-plus-ten-participant-category-plus-three-cap-and-no-duplication pairing layer — six recommendation directions, ten participant unit categories, a single-unit cap of three recommendation items, and open-source-community-governance explicitly named as one of the security-governance sub-topics. This is the CAICT-institute institutional object this series’ prior briefings documented at the technology-standard layer (2026-04-29 ATH protocol dual-open-source release) and at the capability-assessment layer (2026-06/07 first batch of 13 enterprises passing) — now observed at the policy-framework-plus-technology-standard-plus-capability-assessment-plus-governance-practice institutional-chain-completion layer.
  • Tianjin Binhai Xinchuang Industry Bureau (自主可控新鲜事 2026-10-12 report) documents the Tianjin-Binhai-local-government-governance institutional object at the ocean-economy-military-civilian-fusion-to-xinchuang-plus-investment-bureau-to-industry-bureau two-step-institutional-formation layer simultaneously with the administrative-bureau-plus-comprehensive-party-committee-plus-industry-alliance three-layer-architecture layer — a two-year progressive renaming (2021 to 2023) with no public launch ceremony, layered with a comprehensive Party committee absorbing Party-member technical backbones from research institutes and leading enterprises, plus a national taxonomy that places Tianjin as the most institutionalized of four nationwide Xinchuang governance institutional forms, plus a scale metric of 85 billion yuan 2025 Xinchuang industry revenue and 41 new landed enterprises in 2026 Q1. This is the Xinchuang-governance institutional surface this series’ October 11 briefing documented at the SASAC-46-lifetime-reckoning-and-Xinchuang-cobra-effect-policy-tightening layer — now observed at the local-government-two-step-rename-plus-three-layer-architecture-and-most-institutionalized-national-form layer.

The two stories together document that two institutional objects this series’ prior October briefings have documented at the technology-standard-plus-capability-assessment-plus-Xinchuang-policy-tightening layer are now being operationalized at two new institutional layers simultaneously:

  • CAICT Institute moves from technology-standard-plus-capability-assessment (2026-04-29 ATH protocol plus 2026-06/07 first-assessment 13-firms) to policy-framework-plus-technology-standard-plus-capability-assessment-plus-governance-practice institutional-chain-completion — the CAICT institute is not being operationalized at the technology-standard-only layer or the capability-assessment-only layer, but at a four-layer-simultaneous institutional-chain-completion layer in which the policy-framework (2026-05-08 agent guidelines jointly issued by three agencies) plus the technology-standard (2026-04-29 ATH protocol) plus the capability-assessment (2026-06/07 13-firms-passing) plus the governance-practice (2026-10-12 application activity) are four institutionally distinct but related surfaces on the same CAICT-institute-surface.
  • Tianjin Binhai Xinchuang Governance moves from the state-owned-enterprise-and-Xinchuang-policy-tightening layer (this series’ October 11 briefing) to two-step-rename-plus-three-layer-architecture-and-most-institutionalized-national-form — Tianjin is not being operationalized at a single-institutional-form layer (a bureau, or a working group, or a service center, or a department-coordination) but at a three-layer-institutional-architecture layer in which the administrative-bureau form, the comprehensive-party-committee form, and the industry-alliance form are three institutionally distinct but related forms layered on the same local-government Xinchuang governance surface.

One structural observation across both stories.

The structural observation, reading the two stories together, is that both surfaces are being operationalized at institutional layers this series’ prior briefings have documented at the technology-standard-and-policy-tightening layer but not at the institutional-chain-completion-plus-three-layer-architecture layer. This series’ prior five October briefings (October 7, October 8, October 9, October 10, October 11, October 12) documented the pre-IPO-capital-formation institutional object (DeepSeek, Moonshot AI, HKIC-BAAI) at the sovereign-capital-plus-voting-rights-plus-CITIC-Securities-STAR-IPO institutional layer, the Western-foundation-in-China object (PyTorch Foundation, PyTorch Conference China 2026) at the inaugural-co-location-plus-Chinese-Platinum-cohort-majority institutional layer, and the Xinchuang-governance object (openEuler 2026-09 operations report, SASAC Decree No. 46) at the emergent-ecosystem-operations-plus-lifetime-reckoning-plus-cobra-effect-policy-tightening institutional layer. This cycle’s briefing documents that the same state-adjacent-research-institute-and-local-government-institutional-formation institutional pattern is now being operationalized at the CAICT-institute-plus-Tianjin-Binhai Xinchuang-governance institutional layer — two institutional objects that this series has tracked throughout its 52 prior briefings but which have now crossed a first-documented threshold — from the technology-standard-and-policy-tightening layer to two new institutional layers (four-layer-institutional-chain-completion-and-three-layer-architecture).

The institutional-economics reading, in this series’ standard framing, is that the two new layers — four-layer-institutional-chain-completion on the CAICT-institute-surface and two-step-rename-plus-three-layer-architecture on the Tianjin-Binhai-Xinchuang-governance-surface — are not the same institutional form as the technology-standard-and-policy-tightening form. They are different institutional surfaces of the same institutional pattern: the Chinese state is operationalizing the state-adjacent-research-institute-and-local-government-institutional-formation pattern (CAICT-institute-agent-governance four-layer-chain-completion, Tianjin-Binhai Xinchuang governance two-step-rename-and-three-layer-architecture) at the state-adjacent-research-institute-agent-governance-and-local-government-Xinchuang-governance institutional layer.

The two stories together also document an inverse symmetry: the CAICT-institute layer is characterized by institutional-chain-visibility (four layers documented across 14 months of policy-framework, technology-standard, capability-assessment, and governance-practice events) while the Tianjin-Binhai layer is characterized by institutional-formation-invisibility (two-year progressive renaming with no public launch ceremony) — one surface is institutional-chain-visible and one surface is institutional-formation-invisible, but both are being operationalized at the same state-adjacent-research-institute-and-local-government-institutional-formation institutional pattern.

One perspective, not a verdict.

All two stories — CAICT’s 2026-10-12 Internet Agent Governance Innovation Application Practice Activity launch and the 自主可控新鲜事 outlet’s 2026-10-12 report on the Tianjin Binhai Xinchuang Industry Bureau — are best read as observations of institutional movement in progress, not as verdicts on institutional direction. The CAICT four-layer-institutional-chain-completion layer does not guarantee that the CAICT-agent-governance-institutional-chain will remain a policy-framework-plus-technology-standard-plus-capability-assessment-plus-governance-practice form (it may bifurcate, defer, or resolve to a different institutional form as the 2026-10-12 activity moves through its submission-and-review-and-results-release cycle); the Tianjin-Binhai two-step-rename-plus-three-layer-architecture layer does not guarantee that the local Chinese Xinchuang governance institutional surface will remain a full-fledged-bureau-plus-comprehensive-party-committee-plus-industry-alliance form (it may evolve to a more-market-facing service-center form, or to a more-working-group-oriented coordination form, as other local governments document their own Xinchuang governance institutional formations); the 85-billion-yuan 2025 Xinchuang industry revenue and 41-new-landed-2026-Q1 metrics do not guarantee that the Tianjin Xinchuang industrial scale will continue to outpace the Xinchuang institutional formation visibility (the two-step-rename-institutional-formation layer may become institutional-chain-visible if future procurement notices or appointment notices generate dedicated press releases). What this cycle’s briefing documents is that the CAICT-institute-agent-governance surface and the Tianjin-Binhai-Xinchuang-governance surface — two institutional objects this series has tracked throughout its 52 prior briefings — have crossed a first-documented threshold — from the technology-standard-and-policy-tightening layer to two new institutional layers (four-layer-institutional-chain-completion and three-layer-architecture) — and that the two new layers sit on the same unsolved institutional pattern: the state-adjacent-research-institute-and-local-government-institutional-formation object activating first-documented-institutional-chain-completion-and-three-layer-architecture institutional identities this series’ prior fifty-two briefings have not documented at before.

Editorial note on perspective: This briefing presents one institutional-economics reading of Chinese open-source developments, not a verdict. The “institution” in these stories — the CAICT-institute-agent-governance four-layer-institutional-chain-completion layer, the CAICT Cloud-and-Big-Data-Institute-Open-Source-and-Intelligent-Safety-Department hosting layer, the open-source-community-governance-explicit-citation-plus-ten-participant-category-plus-three-cap-and-no-duplication recommendation-design layer, the Tianjin-Binhai-two-step-rename-ocean-military-civilian-fusion-to-Xinchuang-and-investment-bureau-to-industry-bureau institutional-formation layer, the administrative-bureau-plus-comprehensive-party-committee-plus-industry-alliance three-layer-architecture layer, the four-nationwide-category-taxonomy-plus-Tianjin-most-institutionalized-form national-taxonomy layer, the 85-billion-2025-plus-41-new-landed-2026-Q1 industrial-scale-plus-institutional-formation-invisibility two-surface-simultaneous layer — is treated as an object of observation, not a target of critique. The Great Divergence 2.0 framework (FLOSS vs. State-Chartered Codebase vs. Intranet Shared Source vs. Cyber-Estate) is a lens, not a universal answer. One perspective, not a verdict.