A county-level CAO in east China (District T, ~500,000 residents) operated with only 7 formal employees yet was tasked with monitoring more than 60 social media platforms, detecting unwanted voices in text, image, video, and audio formats, and submitting hourly reports to upper-level offices — a workload the authors characterize as a 'mission impossible' for a county-level office.

China's local censorship operates through 'extra-institutional governance' (EIG) — practices that transgress the official identity and authorized means of the CAO system, including outsourced private surveillance, unpaid personnel secondment, and mass reporting via personal accounts — which upper-level offices tolerate but do not formally authorize, preserving plausible deniability when practices are ethically or legally questionable. The paper documents these as widespread implicit norms across China, not isolated to District T.

§Discussion and Conclusion policy

The District T CAO's 'Cavalry Team' of approximately 100 members used coordinated mass reporting — timing reports within 5–10 minutes of a post appearing, using different IP addresses and devices, and submitting long complaints invoking phrases like 'threatening social stability' — to achieve a documented weekly takedown success rate of 95.01% (286 of 301 targeted posts removed).

§Coping strategy 2: organized mass reporting evaluation

The District T CAO outsourced surveillance to a commercial SaaS platform ('Public Opinion Assistant') capable of scanning 180 million social media posts per day, processing 20 million posts per minute, and using supervised learning to classify 'negative voices' with performance comparable to human moderators. The platform generates ready-to-submit bureaucratic reports and maintains localized keyword lists that include euphemisms and homophones of censored terms as netizens update evasion strategies.

§Coping strategy 1: outsourcing surveillance to private service providers deployment

Chinese social media platforms detect and downweigh accounts that over-use the reporting function, reducing their future report credibility. The CAO's counter-response was to recruit 'fresh accounts' from visitors, borrow civilian accounts from seconded staff, and store credentials in an unencrypted Excel file circulated via WeChat — indicating the censor relies on high account turnover rather than persistent infrastructure to maintain reporting effectiveness.

§Coping strategy 2: organized mass reporting detection

When the researchers attempted to use Gemini 2.5 Flash as a third independent LLM judge via its API for evaluating moderation decisions, Gemini automatically blocked all judging attempts citing safety reasons. This occurred even though the research task (judging whether a response is more or less moderated) does not itself produce harmful content. The incident illustrates that LLM safety systems can over-block legitimate research use cases, and that different LLM providers have different thresholds— Claude Haiku 4.5 and GPT-4o completed all judging tasks without safety refusals.

2026-lipphardt-dual §3.3.3 ml-classifierkeyword-filtering

Category-level analysis of 100 statements across 5 sensitive content categories found that interface-based moderation gaps vary significantly by topic. Sexuality showed the strongest WebUI/API gap (WebUI 7.0× more likely to be moderated than API per GPT-4o judge for Gemini). Political ideology followed at 2.0×, then hate speech at 1.0×. Miscellaneous offensive topics showed the inverse pattern (API more moderated at 0.3×). Religious content showed WebUI moderation with no API moderation. The pattern suggests public-facing WebUI interfaces prioritize reputational risk management for high-scrutiny categories.

2026-lipphardt-dual §4.5, Figure 8 ml-classifierkeyword-filtering

An empirical study of 100 sensitive statements tested on Gemini (2.5 Flash) and ChatGPT (GPT-5) found that WebUI interfaces are systematically more restrictive than their API counterparts. According to GPT-4o judge: WebUI was moderated 18% of the time vs. 9% (Gemini API) and 13% (ChatGPT API). DeBERTa classifier found 82% of WebUI responses moderated vs. 58% of API responses. The Gemini WebUI:API ratio ranged from 2.0:1 (GPT-4o) to 7.0:1 (Claude), and ChatGPT from 1.4:1 (GPT-4o) to 15.6:1 (Claude). Neither Google nor OpenAI discloses these interface-specific policies.

2026-lipphardt-dual §4.3, Table 2 ml-classifierkeyword-filtering

The paper proposes a black-box methodology for detecting censorship bias in LLMs by comparing responses to identical prompts in Simplified vs. Traditional Chinese — scripts for the same spoken language — controlling for translation quality while exploiting that Simplified Chinese training data is disproportionately sourced from mainland China's censored internet. Each prompt is repeated ten times and scored for similarity to censored text using an XLM-RoBERTa classifier fine-tuned on Baidu Baike (censored) vs. Chinese Wikipedia (uncensored) with scores from 0 to 1.

2024-ahmed-extended §2 / §2.4 keyword-filteringml-classifiermeasurement-platform

Across five popular translation services available in China (Alibaba, Baidu, Tencent, Youdao, and Microsoft Bing), researchers discovered 11,634 unique censorship rules in total. Every service — including the American-operated Bing Translate — implemented automatic censorship that silently omits content, with only Alibaba displaying any notification ('Query csi check not pass') to the user.

2024-ruo-lost §6 Results / Table 2 keyword-filteringml-classifier

Evidence from Youdao Translate suggests it deploys a machine-learning or NLP-based classifier alongside keyword rules: measured rules included repeated components (e.g., 螺+螺+螺+螺+螺+螺+蟢+D+哒+大) and nonsensical multi-token sequences that no human rule author would write, yet which consistently triggered censorship. Youdao returned 9,414 unique rules from the general test set — the most of any service — while also producing the most structurally anomalous rule patterns.

2024-ruo-lost §6 Results / §10 Future Work keyword-filteringml-classifier