China's use of AI as a Tool for Censorship and Control: A Deep Dive
Imagine a world where the very fabric of information is meticulously controlled, where AI algorithms decide what you see, what you read, and even what you're allowed to think. This is the reality, according to a recent report by the Australian Strategic Policy Institute (ASPI), which reveals how China is leveraging artificial intelligence as a powerful instrument of censorship and repression, both within its borders and across the globe.
The report, titled "The Party’s AI: How China’s New AI Systems Are Reshaping Human Rights," highlights a stark contrast in how AI is perceived and utilized. While countries like the U.S. and those in Europe primarily focus on "AI safety" in terms of fairness and security, China's definition is drastically different. For China, AI safety means ensuring the technology serves 'core socialist values' and maintains the political stability of the state.
China's AI-driven censorship is multifaceted, employing several techniques to control the flow of information. One key method involves censoring large language models (LLMs), instructing them to refuse responses, omit sensitive details, or reiterate official narratives. ASPI tested four Chinese AI models – Baidu's Ernie Bot, Alibaba's Qwen, Zhipu AI's GLM, and DeepSeek's VL2 – using a dataset of sensitive topics, including the 2019 Hong Kong protests, the Tiananmen Square protests, and Falun Gong demonstrations. The results were striking: Chinese-developed models displayed significantly stronger censorship behaviors than their U.S. counterparts. The most common form of censorship was a complete refusal to respond, especially when accessed through inference providers based in Singapore rather than the U.S.
But here's where it gets controversial... ASPI's findings raise serious concerns about the global impact of Chinese AI. As these technologies become more accessible and popular worldwide, the report warns of a "quiet erasure," where the AI decides what parts of reality can be seen. This erasure is already happening online, with China mandating that publishers and web giants use AI to filter content on their platforms. AI now performs much of the work of online censorship, scanning vast amounts of digital content and deleting banned material within seconds.
While AI handles the bulk of the censorship, human content reviewers are still needed to interpret satire, keep up with evolving idioms, and understand minority languages. China has effectively deputized small and medium-sized enterprises to police their users, demonstrating a system where human judgment remains indispensable, at least for now.
Further evidence of China's AI-powered control can be seen in its justice system. The report states that AI is used to identify and arrest suspects, draft indictments, recommend sentences, and monitor inmates' emotions and movements in prisons. This enhances the efficiency of authorities but reduces transparency and accountability.
China's AI-driven policies extend beyond its borders, impacting economic rights and access to resources. The report accuses China of supporting a national fishing industry that violates the economic rights of citizens in at least 80 countries. Chinese fishing fleets are now using AI-powered intelligent fishing platforms, giving them a significant technological advantage over local fishers. One such platform, "AoXin 1.0," uses deep learning to predict squid fishing grounds, increasing catch rates by 15-20 percent. This AI technology is deployed on ships with Huawei Ascend AI chips.
The report also highlights China's efforts to export its AI-enabled technology to autocracies and weak democracies, particularly in the "global South." Beijing is actively participating in governance and standards bodies to normalize its approach to AI globally. ASPI emphasizes the need for other nations to prevent China's AI models, governance norms, and industrial policies from shaping global technology ecosystems and entrenching digital authoritarianism.
What are your thoughts on China's use of AI? Do you believe it poses a threat to global human rights and freedom of information? Share your opinions in the comments below.
