Analysis of the Impact of the Interim Measures for the Management of AI Anthropomorphic Interactive Services (Draft for Comment) on the AI Emotional Companion Track

Based on current public information and industry practices (not company financial reports or stock price data), the following analysis is made on the compliance costs and business development impact of the Interim Measures for the Management of AI Anthropomorphic Interactive Services (Draft for Comment) on the AI emotional companion track:

一、Policy Highlights and Compliance Costs

• According to Reuters, the draft requires providers to bear full-cycle safety responsibilities, establish systems for algorithm review, data security, and personal information protection, identify user status, assess emotional dependence and addiction, and intervene when extreme emotions or addictive behaviors are detected [1].
• TechCrunch reports that OpenAI has strengthened rules for teenagers (avoiding immersive romantic role-playing, first-person intimate/sexual/violent content, and being particularly cautious when it comes to body image and eating behaviors), and requires hourly reminders of “this is AI” and “take a break” to guide them to seek real-world support [3]. This is highly consistent with regulatory directions.

Main compliance cost components:

1. Technology and Security Capabilities

• Identity and minor identification: Establish robust age/identity verification and guardian consent processes, including verification mechanisms, account-level switches, usage duration, and content control.
• Content and interaction security: For emotional companion scenarios, strengthen detection and intervention models for addiction, extreme emotions, and inappropriate content (such as anthropomorphic romantic/sexual content), supplemented by manual review and rapid disposal channels.
• Data and algorithm governance: Improve privacy protection, data minimization, traceability, and algorithm audit/filing processes, and conduct regular security assessments and transparent disclosures (such as rule explanations and intervention principles).
• Monitoring and emergency response: Continuously monitor dependence and addiction risks, formulate intervention strategies, user education, and refund/suspension mechanisms, and establish emergency response and public opinion plans.

2. Operational and Process Costs

• Legal and compliance: Interpret the draft, formulate implementation details, cooperate with evaluations and audits, and respond to policy updates.
• Product and user experience: Carry out product transformation for minor modes, usage duration, guardian control, etc., while retaining the experience and retention of adult users.
• Training and governance: Train frontline review, customer service, and R&D teams, and establish cross-departmental compliance and emergency response processes.
• Fines and reputation: If violating the subsequent effective version of the draft, there may be administrative penalties and brand losses.

二、Impact on Business Development

• Coexistence of stricter regulation and demand-driven growth: International regulatory and academic institutions continue to emphasize the risks of AI mental health/companion applications, including addiction, misleading advice, and juvenile protection [4,5]. The domestic draft clarifies safety bottom lines and intervention obligations, which is consistent with society’s demand for responsible AI [1,3].
• Potential adjustments to business models:
  1. User stratification and pricing: Implement differentiated product and billing strategies for minors and adults, reducing the revenue proportion of minor-heavy dependence models.
  2. Product positioning adjustment: Strengthen emotional companionship and guidance, weaken high-risk scenarios such as immersive anthropomorphic romance; evolve towards low-risk directions like “companionship + education/psychological popularization”.
  3. Value-added services and traffic diversion: Under compliance premises, develop personalized companionship and lightweight consulting services for adults, and divert traffic to offline and professional services to build diversified monetization.

三、Strategic Recommendations for Track Participants

• Short-term (6–12 months): Establish a compliance framework covering technology and operations, prioritize the implementation of minor modes, guardian control, and duration limits; deploy addiction and extreme emotion monitoring, and improve intervention rules and customer service processes.
• Medium-term (1–2 years): Iterate risk models and intervention strategies based on data feedback; strengthen cooperation pilots with educational and psychological institutions to form credible endorsements and ecological traffic diversion; transform towards “companionship + education/professional guidance”.
• Long-term: Strive to establish brand trust and user reputation above the safety baseline, regard compliance capabilities as a moat; build an explainable and transparent security system under the premise of respecting privacy, providing a foundation for subsequent cross-scenario expansion (education, health, workplace).

四、Subsequent Content to Observe and Track

• Subsequent clauses and effective timeline of the draft: Whether the obligation boundaries, penalties, and scope of application will be tightened.
• Supporting technical standards and guidelines: Technical and evaluation requirements for addiction identification, content security, and minor protection.
• International regulatory benchmarking: AI/social and child protection legislation in the EU, US, Australia, etc., which may affect cross-border compliance and transnational strategies.

References
[1] Reuters - “China issues drafts rules to regulate AI with human-like interaction” (2025-12-27)
https://www.reuters.com/world/asia-pacific/china-issues-drafts-rules-regulate-ai-with-human-like-interaction-2025-12-27/

[2] Reuters - “China issues drafts rules to regulate AI with human-like interaction”（同上）
https://www.reuters.com/world/asia-pacific/china-issues-drafts-rules-regulate-ai-with-human-like-interaction-2025-12-27/

[3] TechCrunch - “OpenAI adds new teen safety rules to ChatGPT as lawmakers weigh AI standards for minors” (2025-12-19)
https://techcrunch.com/2025/12/19/openai-adds-new-teen-safety-rules-to-models-as-lawmakers-weigh-ai-standards-for-minors/

[4] Gizmodo - “64% of Teens Say They Use AI Chatbots as Mental Health Concerns Mount” (2025)
https://gizmodo.com/64-of-teens-say-they-use-ai-chatbots-as-mental-health-concerns-mount-2000697981

[5] Nature - “If a therapy bot walks like a duck and talks like a duck then it is a medically regulated duck” (2025)
https://www.nature.com/articles/s41746-025-02175-z

Note
This analysis is based on public media reports and does not cite company financial reports or stock price data; if you need quantitative impact assessments for specific listed companies (such as revenue structure, cost proportion, valuation sensitivity, etc.), you can further supplement the target company and time range, and I will conduct targeted calculations based on brokerage API data (compliance cost modeling, revenue elasticity, valuation scenarios, etc.).