AI Companions Face Shutdown: Doubao and Tongyi Qianwen Remove Personification Features
The clock is ticking for AI companions in China. As the July 15 enforcement date for the Interim Measures for the Management of Human-like Interactive Services of Artificial Intelligence approaches, major tech players are scrambling to comply. ByteDance's Doubao and Alibaba's Tongyi Qianwen have announced they will disable customizable human-like agent features, becoming the first major platforms to adjust their products before the new rules kick in.
Doubao sent a notice to users on Friday evening, stating that its agent function will be discontinued starting July 15 due to "product function adjustments." The notice added that data related to this feature will be handled according to the company's privacy policy starting October 15, and users will no longer be able to view or recover related content within the app. Tongyi Qianwen is moving even faster—its "human-like interactive agents and user-created agent functions" will be disabled on July 10, with the broader "Tongyi agent functions and services" fully discontinued by July 15. After that, users won't be able to access existing agent settings or historical conversation records.

New Regulations Target Emotional Bonds
Both apps previously offered agent pools where users could create AI assistants with fixed personas and tones—think named assistants, learning tutors, role-playing characters, or emotional companions. The timing of these changes is no coincidence. The Interim Measures, released in April, explicitly regulate services that "simulate human personality traits, thinking patterns, and communication methods, and provide continuous emotional interaction." However, they exclude customer service bots, knowledge Q&A tools, workplace assistants, and educational tools. The regulations cite risks like the spread of extreme ideas, privacy leaks, harm to users' mental health, and potential addiction.
It's worth noting that Tencent had already removed similar agent features from its AI assistant Yuanbao in June. So far, the three tech giants in China are collectively stepping back from human-like agent functions. The message from regulators is clear: AI agents should be part of productive infrastructure, not emotional crutches. Between May and June, regulators also issued guidance on orderly AI agent development and national standards for interoperability, covering identification, authorization, connectivity, and traceability.
User Backlash: Emotional Support Disappears Overnight
The policy shift has sparked backlash. A Weibo user @Doubao official account lamented, "Why are we discontinuing the agent? They have always been our emotional support." The user complained that long-term chat records and emotional connections are hard to export or transfer, and the lack of a smooth data migration path makes the adjustment feel abrupt. When AI companions become a spiritual reliance for some users, the balance between regulation and product design clearly needs more buffer and transition time.
Key Points:
- Doubao and Tongyi Qianwen are disabling AI personification features ahead of July 15 regulations.
- Tencent's Yuanbao had already removed similar features in June.
- New rules target continuous emotional interaction, citing risks like addiction and privacy leaks.
- Users express frustration over losing emotional support and difficulty migrating data.