Summary
ByteDance's Doubao and Alibaba's Qwen are disabling custom AI agent features to comply with China's new Interim Measures on anthropomorphic AI interaction services, effective July 15. The regulations target sustained emotional interaction services while sparing workplace productivity tools.
Major AI Platforms Simultaneously Remove Custom Agent Features
China's consumer AI application market is undergoing a significant regulatory adjustment. According to the South China Morning Post, ByteDance's Doubao and Alibaba's Qwen, two of China's largest consumer AI applications, are disabling their custom agent features in response to upcoming regulations.
Doubao notified users on Friday, July 4, that its agent feature would go offline on July 15, citing "product function adjustments." Related data will stop being viewable or recoverable within the app after October 15. Qwen followed on Saturday, announcing that humanlike interactive agents and user-created agents would be disabled on July 10, with broader agent functions going offline by July 15. Users will lose access to agent settings and their previous conversations.
Tencent had already taken preemptive action in June, removing a similar feature from its Yuanbao assistant. Chinese state media confirms that these removals are driven by regulatory compliance requirements.
These applications previously allowed users to build named assistants, tutors, role-playing characters, or companions with fixed personas and speaking styles. The removal of these features marks a significant shift in how consumer AI services can be offered in China.
New Regulations Target "Anthropomorphic Emotional Interaction" Services
These changes stem from the Interim Measures for AI Anthropomorphic Interaction Services issued in April by the Cyberspace Administration of China and four other agencies. The regulations take effect on July 15, 2026.
The measures specifically target services that "simulate human appearance, voice, emotions, or behavioral characteristics," particularly those offering sustained emotional interaction. The regulatory framework distinguishes between different types of AI agents, with the primary focus on companion-style applications that may foster emotional dependency.
Notably, the regulations adopt a differentiated approach. Workplace and productivity tool agents are expected to remain unaffected, with enforcement concentrated on companion applications that could lead to emotional attachment or dependency. This distinction reflects regulators' concerns about the psychological and social impacts of AI services that blur the line between human and machine interaction.
User reactions have been mixed. Some users express concern about losing "AI girlfriend" and other emotional companion features, particularly regarding the loss of chat histories. Others view the regulations as a necessary measure to prevent excessive emotional dependency on AI systems.
The data preservation timeline has become a particular point of contention. Users have only until October 15 to access or export their conversation histories, after which this data will become permanently inaccessible. This relatively short window has prompted criticism from users who view their AI interactions as valuable personal records.
AI Tool Security Scrutiny Intensifies Across the Board
Regulatory tightening extends beyond consumer applications. According to multiple reports, Alibaba will ban employees from using Anthropic's programming tool Claude Code starting July 10, classifying it as high-risk software and instructing employees to use the company's own Qoder tool instead.
Anthropic already prohibits Chinese companies, as well as foreign entities owned by those companies, from using its models. The company has reportedly been working to close loopholes that allow Chinese users to access Claude. According to a recent Reddit post, some loophole-closing efforts involved a version of Claude Code that could secretly identify Chinese users.
Anthropic's Thariq Shihipar explained on X that this was "an experiment we launched in March that was meant to prevent account abuse from unauthorized resellers and protect against distillation." Distillation refers to the practice where AI models are trained on the outputs of other models. Shihipar noted that "the team has landed stronger mitigations since then and we've actually been meaning to take this down for a while."
Despite Anthropic's own restrictions and mitigation efforts, Alibaba's decision to classify Claude Code as high-risk and ban its use reflects heightened corporate security concerns. The move highlights how AI programming tools are facing increasingly rigorous security reviews in enterprise environments, particularly regarding potential backdoor risks and data security issues.
This incident occurred shortly after an AMD AI director's criticism raised alerts among Chinese tech companies, prompting Alibaba to take preemptive action. The timing underscores how quickly security concerns can escalate in the current regulatory environment.
Industry Implications of Stricter Regulations
China's tightening of AI application regulations reflects the government's effort to balance technological innovation with social governance. For the AI industry, this means product design and feature development must incorporate compliance considerations at much earlier stages.
From a technical architecture perspective, major AI platforms need to establish more granular feature classification and risk assessment mechanisms. Emotional companion features must be clearly distinguished from productivity tools to accommodate differentiated regulatory requirements. This may drive the industry toward more specialized, vertical development rather than pursuing comprehensive feature integration.
For users, the removal of custom agent features means personalized AI experiences will be limited. However, from another perspective, this may encourage developers to concentrate resources on improving AI capabilities in practical scenarios such as work and learning, rather than focusing on emotional interaction simulation.
The strengthening of security reviews also presents new challenges for enterprise AI applications. Alibaba's ban on Claude Code demonstrates that even internationally renowned AI tools must undergo rigorous security assessments before entering corporate environments. This may promote the development of domestic AI tools, though care must be taken to avoid innovation constraints from excessive protectionism.
The regulatory approach also raises questions about data sovereignty and AI model governance. As companies navigate compliance requirements, they must balance between using best-in-class global tools and adhering to local security standards. This tension is likely to shape product development strategies and corporate AI adoption patterns in the coming years.
A Microcosm of Global AI Regulatory Trends
China's AI regulatory tightening is not an isolated phenomenon but part of a global trend toward strengthened AI governance. The EU's AI Act and various U.S. state AI regulatory proposals all reflect government concerns about the potential risks of AI technology.
However, China's regulatory approach differs in its emphasis on content and social impact, particularly focusing on AI's potential effects on user emotions and behavior. This contrasts with Western regulations that tend to focus more on technical aspects such as privacy, bias, and transparency.
For the global AI industry, divergent regulatory frameworks across jurisdictions mean products require localization for different markets. This involves not just language and cultural adaptation, but deeper compliance requirements affecting feature design, data processing, and security mechanisms.
The regulatory landscape also highlights different philosophical approaches to AI governance. China's emphasis on emotional interaction and social stability reflects distinct policy priorities compared to the West's focus on individual rights and algorithmic accountability. These differences will likely persist and potentially widen as AI capabilities advance.
As AI technology continues to evolve, regulatory frameworks will undergo constant adjustment. Finding the balance between promoting innovation and protecting users will be a long-term challenge faced jointly by governments and the AI industry worldwide. The Chinese case offers insights into how regulations can rapidly reshape product offerings and user experiences, providing lessons for other jurisdictions considering similar measures.
The current regulatory moment also serves as a reminder that AI development cannot proceed independently of social and political contexts. As AI systems become more sophisticated and integrated into daily life, the questions they raise about human-machine interaction, emotional dependency, and social norms will only become more pressing. How different societies choose to address these questions through regulation will significantly influence the trajectory of AI development globally.
Source: link