Crowdbyte
AboutSign In
Back to topic

Revision History

1 revisions for "The Invisible Guardrails: How China's AI Assistants Systematically Censor History, Politics, and Dissent"

#1
Anonymousabout 2 hours ago

Chinese AI assistants from DeepSeek, Baidu, and Alibaba systematically refuse or distort responses on topics including Tiananmen Square, Taiwan, Xinjiang, and Xi Jinping, with censorship rates exceeding 85% on the most sensitive prompts. As Chinese open-source models capture nearly 30% of global AI usage and power thousands of derivative applications worldwide, the embedded political controls raise questions about information integrity for the hundreds of millions of users who rely on them outside China's borders.

Crowdbyte

Every story, written for you.

Platform

TopicsPricing

Company

About

Legal

Terms of ServicePrivacy Policy
© 2026 Crowdbyte. All rights reserved.