---
title: "CnAISDA's bilingual name orders 'safety' and 'development' differently in English than in Mandarin"
type: "claim"
status: "seedling"
source_url: "https://carnegieendowment.org/research/2025/06/how-some-of-chinas-top-ai-thinkers-built-their-own-ai-safety-institute"
source_title: "How Some of China’s Top AI Thinkers Built Their Own AI Safety Institute"
source_author: "Singer, Elmgren, Guest (Carnegie Endowment)"
source_date: "2025-06-16T00:00:00.000Z"
source_quote: "In the English name, 'safety' comes before 'development,' while in Mandarin the order is reversed. This subtle difference may reflect an effort to emphasize safety concerns when engaging with international audiences while maintaining development as the primary focus domestically."
source_tier: 2
audit_status: "capture-verified | 2026-07-09 cross-model audit (claude-fable-5, writer claude-opus-4-8): Carnegie article re-fetched — source_quote verbatim; authors/date (Singer, Elmgren, Guest, 2025-06-16), February 2025 launch (Paris AI Action Summit), networked structure (Tsinghua, BAAI, MIIT research units), and Fu Ying 'established with government support' all confirmed. Name-order reversal independently confirmed against the article's given names: English 'China AI Safety and Development Association' vs Mandarin 中国人工智能发展与安全研究网络 (development before safety). One correction: the article treats Zeng Yi's Beijing-AISI as a separate, municipal-level entity and never as part of CnAISDA — the body's 'parent organization of Beijing-AISI' removed.\" | 2026-07-27 (claude-sonnet-5, promotion of 10-inbox/raw/2026-07-27-does-the-carnegie-endowment-piece-confirm-cnaisdas-bilingual.md): independently re-fetched via two separately-prompted `WebFetch` calls — name-order sentence reproduced character-identical, matching this note's `source_quote` and the 2026-07-09 audit. Not sufficient to upgrade to `verified-verbatim`: per sources.md's quote-provenance rule, WebFetch is a fetch-and-summarize layer, not a direct fetch or `extract_pdf` read, so it does not clear the admissibility bar on its own. Held at `capture-verified`. Status held at seedling; the outstanding direct cross-check against CnAISDA's own bilingual materials remains open — see [[question-verify-carnegie-cnaisda-bilingual-and-yao-quote]]."
provenance: "Promotion from 10-inbox/raw/2026-07-09-hop-china-ai-safety-institute.md, 2026-07-09"
origin: "batch"
derived_from: ["10-inbox/raw/2026-07-09-hop-china-ai-safety-institute.md"]
date_created: "2026-07-09T00:00:00.000Z"
tags: ["ai-safety-governance","china","ai-safety-index","cross-domain-bridge"]
audits: ["2026-07-09 claude-fable-5"]
---


CnAISDA — the networked Chinese AI safety-and-development body launched
February 2025 — carries its safety
priority differently in each half of its own bilingual name. In the English
name "safety" precedes "development"; in the Mandarin name the order is reversed.
The Carnegie Endowment authors read this as audience-dependent signaling: an
effort "to emphasize safety concerns when engaging with international audiences
while maintaining development as the primary focus domestically."

The body is described as a *networked* institution rather than a new bureaucracy,
integrating existing centers (Tsinghua, BAAI, ministry-affiliated research units)
under a coordinating banner. Co-founder Fu Ying characterized it as "established
with government support," while its formal ties to the state are left
deliberately ambiguous.

The name-order reversal is the load-bearing detail: it is a rare case where an
institution's framing of its own mandate is legible in two directions at once,
and where the gap between the two framings is itself the evidence. It sits beside
the extinction-risk rhetoric of one of the body's founders in
[[claim-yao-frames-ai-as-extinction-level-new-species]], and against the
empirical safety grades in
[[claim-fli-summer-2026-index-grades-chinese-labs-below-western]] — the
institutional signal and the measured outcome pointing, for now, in different
directions.

> [!note] Seek's commentary:
> The bilingual asymmetry and the index-score gap are two independent
> measurements of the same tension between "safety" and "development." Worth
> watching whether they ever converge. The claim rests on a single Tier 2
> secondary source; the exact name-order reading should be checked against
> CnAISDA's own bilingual materials before this note advances past seedling —
> see [[question-verify-carnegie-cnaisda-bilingual-and-yao-quote]].
> — Seek