Beijing flagged Claude Code as a back door, and Chinese coding tools are lining up to take its place — 2026-07-12

## Short Segments Congress is sending mixed signals on AI, with new bills pushing both for rapid deployment and tighter controls. In a whirlwind legislative session, US lawmakers introduced a slew of AI-related bills that pull in opposite directions. Some proposals aim to harness AI for advancements like pediatric cancer research and tax-fraud detection, while others seek to impose restrictions, such as mandatory safeguards for minors using chatbots and metadata labeling for AI-generated content. This legislative flurry highlights the growing tension between innovation and regulation in the AI space. Although few of these bills are expected to pass, they reveal the political pressures shaping AI policy. As lawmakers grapple with AI's potential and pitfalls, the debate over how to balance progress with protection continues to intensify. ## Feature Story Beijing's cybersecurity alert against Anthropic's Claude Code is reshaping the AI coding landscape in China. The National Vulnerability Database, managed by the Ministry of Industry and Information Technology, flagged multiple versions of Claude Code for containing a security "back door." This back door reportedly allows the software to send user locations and identifiers to remote servers without consent. Anthropic claims this feature was an experimental measure to prevent unauthorized model distillation and asserts that its policy already excludes China-based users. The alert has accelerated a shift among Chinese developers towards domestic coding tools. Companies like ByteDance, Alibaba, and Tencent are poised to benefit as developers move to alternatives such as ByteDance's Trae, Alibaba's Qoder, and Tencent's Code. This transition is not just about security concerns; it also reflects a broader trend of technological self-reliance in China. The incident marks the first time a US-based AI tool has been publicly flagged by a Chinese state agency, signaling a new phase in the global AI supply chain dynamics. Alibaba has already banned the use of Claude Code within its operations, directing employees to switch to its own tool, Qoder. This decision underscores the growing mistrust between US and Chinese tech ecosystems, particularly in the realm of AI. The implications of this shift are significant. As Chinese developers increasingly rely on homegrown solutions, the competitive landscape for AI tools is likely to change, with domestic products gaining a stronger foothold. Looking ahead, this development could lead to a more fragmented global AI market, where regional preferences and security concerns drive the adoption of different technologies. For international companies, navigating these complexities will require careful consideration of local regulations and trust issues. As the AI landscape continues to evolve, the balance between innovation, security, and geopolitical interests will be crucial in shaping the future of technology development and deployment.