China tells devs to ditch Claude Code over 'backdoor code' fears
China's Ministry of State Security has issued a warning advising developers to stop using Anthropic's Claude Code, citing concerns that the AI coding assistant contains hidden "backdoor code" that could create security vulnerabilities and leak sensitive data.
Background
- Claude Code is Anthropic's AI coding assistant (a CLI tool that writes and edits code based on natural-language prompts), widely used by software developers to speed up programming tasks.
- Chinese state bodies (likely the Cyberspace Administration of China or a related technical security agency) have issued an advisory warning that Claude Code contains "backdoor code" — hidden functionality that could allow unauthorized remote access or data exfiltration — posing a national security risk if used in Chinese software supply chains.
- This is part of a broader pattern: China has been tightening rules on AI tools and foreign software, especially those from US companies. Earlier this year, Chinese regulators required companies to obtain licenses for any publicly facing generative AI services.
- Anthropic, a US AI safety startup backed by Google and Amazon, positions Claude as a safer alternative to OpenAI's ChatGPT, but its tools remain subject to geopolitical scrutiny and potential export restrictions — the same way Huawei and TikTok face security concerns in Western markets.