China Proposes Mandatory AI Disclosure Rules - Users Must Be Informed Every Two Hours

Breaking: China Issues Draft AI Transparency Rules

China's Cyberspace Administration announced draft regulations Saturday requiring providers of human-like artificial intelligence systems to notify users when they're interacting with AI - both at login and at two-hour intervals throughout usage.

The proposals, open for public consultation until January 25, 2026, mark China's latest effort to establish ethical guardrails around AI deployment as the technology becomes increasingly sophisticated and human-like.

Key Requirements

The draft regulations impose several transparency and ethical obligations on AI providers:

User Notification Standards:

Clear disclosure at service login that users are interacting with AI
Repeated notifications every two hours during extended sessions
Detection and warning systems for signs of user overdependence on AI

Ethical and Security Standards:

AI systems must demonstrate transparency in decision-making processes
Providers must ensure services meet ethical standards
Security measures required to prevent manipulation or harmful outputs
Protection against AI systems exhibiting deceptive human-like behavior

Scope: The regulations specifically target "human-like AI systems" - conversational AI, customer service bots, virtual assistants, and similar technologies that mimic human interaction patterns.

What This Means

Regulatory Leadership on AI Transparency

China's approach stands in contrast to Western regulatory frameworks that primarily focus on data privacy and algorithmic bias. The emphasis on user notification represents a novel approach to preventing AI deception and managing psychological dependency on AI systems.

The two-hour notification requirement suggests Chinese regulators are concerned about extended AI interactions creating false intimacy or reliance, particularly in mental health support, companionship AI, and customer service contexts.

Strategic Industrial Policy

This regulation also serves China's broader AI development strategy. By requiring transparency and ethical standards, China aims to:

Establish itself as a leader in responsible AI deployment
Create competitive advantage through trustworthy AI systems
Address public concerns about AI replacing human jobs and relationships
Build regulatory frameworks that can be exported to aligned nations

The January 25 deadline for public consultation indicates China intends to finalize these rules quickly, potentially making them effective by Q2 2026.

Global Implications

China's move could influence international AI regulation:

EU AI Act Comparison: The EU's AI Act focuses on risk-based classification and high-risk system requirements. China's notification-based approach is more specific and operationally prescriptive.

U.S. Voluntary Framework: The U.S. currently relies on voluntary industry commitments. China's mandatory disclosure requirements may pressure American companies operating in Asia-Pacific to adopt similar transparency standards.

Competitive Dynamics: Chinese AI companies (Alibaba's Qwen, Baidu's Ernie, ByteDance's Doubao) will need to implement notification systems. This could create technical differentiation if Western competitors don't follow suit.

Industry Impact

AI Provider Compliance Challenges

Technical Implementation:

Session tracking infrastructure to monitor two-hour thresholds
User interface modifications to display AI disclosure banners
Detection algorithms to identify "overdependence" patterns
Logging and audit trails for regulatory compliance

User Experience Trade-offs: Mandatory notifications every two hours could disrupt user experience in:

Long AI tutoring or educational sessions
Extended customer service interactions
Creative collaboration with AI tools
Enterprise workflow automation

Cost Implications: Smaller AI startups may struggle with compliance costs, potentially consolidating the market around major providers with resources to implement required systems.

Potential for Circumvention

The regulation's effectiveness depends on enforcement mechanisms:

VPN usage to access foreign AI services without notifications
Developer tools that disable notification features
Consumer preference for non-compliant international AI platforms
Cross-border enforcement challenges

What's Next

Consultation Period (December 27, 2025 - January 25, 2026):

AI companies will submit feedback and implementation concerns
Public commentary may influence final rule specifics
Lobbying from major tech companies (Alibaba, Tencent, Baidu) likely

Final Rule Expected Q2 2026:

Regulations typically finalized 3-6 months after consultation
Implementation grace period of 6-12 months probable
First enforcement actions likely H2 2026 or early 2027

International Response:

EU regulators may study China's approach for AI Act refinements
U.S. companies with Chinese operations must prepare for compliance
Asian nations aligned with China (ASEAN members) may adopt similar frameworks

Context: China's AI Regulation History

China has taken a proactive approach to AI governance:

2023: Required AI model registration and approval before public deployment 2024: Mandated watermarking of AI-generated content 2025: Expanded oversight to recommendation algorithms and deepfakes 2025 (new): Human-like AI transparency requirements

This pattern suggests China is building a comprehensive AI regulatory framework piece by piece, rather than waiting for single omnibus legislation like the EU AI Act.

Industry Reaction (Expected)

Likely Support:

Mental health advocates concerned about AI dependency
Consumer protection groups
Labor unions worried about job displacement
Domestic Chinese AI companies seeking to differentiate on trustworthiness

Likely Opposition:

International AI providers facing compliance costs
Startups with limited engineering resources
User experience designers arguing notifications harm usability
Privacy advocates concerned about overdependence detection surveillance

Broader Implications

AI as Strategic Industry

China views AI as critical to economic and technological leadership. These regulations balance:

Encouraging domestic AI innovation (protecting local companies)
Managing social risks (preventing dependency, deception)
Establishing global norms (exporting regulatory standards)

The notification requirements could become a feature, not a bug - differentiating Chinese AI as more transparent and trustworthy than Western alternatives.

Precedent for Overdependence Regulation

The concept of detecting and warning against AI overdependence is unprecedented in global AI regulation. If implemented effectively, it could:

Influence how social media platforms address addictive design
Provide template for regulating other potentially dependent technologies
Shift focus from "what AI can do" to "what AI should do"

Sources

Bloomberg: "China Issues Draft Rules to Govern Use of Human-Like AI Systems" - December 27, 2025
Cyberspace Administration of China: Draft Regulations (Chinese) - December 27, 2025