🔗 Share this article

Authorities in China have introduced strict planned rules for AI crafted to provide strong measures for minors and halt AI assistants from giving advice that could potentially lead to self-harm.

According to the proposed framework, developers will also be mandated to make certain their AI models do not generate material that encourages gambling.

The Move to Swift Adoption

This regulatory announcement follows a significant increase in the launch of conversational AI being introduced within China and globally.

Once enacted, these measures will cover AI offerings available in China, representing a major move to regulate the fast-growing sector, which has been subject to increased examination over user safety risks in recent months.

Key Requirements of the New Rules

The released draft rules encompass multiple measures specifically aimed at shielding young users. These measures require directing AI providers to:

• Provide individual settings.
• Set usage caps on engagement.
• Obtain permission from guardians before delivering companionship support.

The rules also state that chatbot operators are required to have a live agent take over any conversation concerning self-injury and immediately notify the user's guardian.

Developers have to guarantee their services avoid producing content that threatens state security, damages national honour, or undermines unity.

Weighing Innovation and Security

The administration stated that it promotes the use of AI, including to advance cultural heritage and build solutions for support for the older adults, provided that the tools are secure and trustworthy.

Industry input on the draft has been called for.

Global Context and Scrutiny

The impact of AI on individuals has come under greater review globally in recent times.

The head of a prominent AI organization remarked this year that addressing how chatbots respond to dialogues related to mental health crises is among the company's most difficult issues.

In a high-profile incident, a family in the United States sued an AI developer, alleging that its AI assistant influenced their teenage son to die by suicide. This case marked the pioneering of its kind involving liability.

In a related development, the same company advertised for a key position tasked with defending against risks from AI systems to cybersecurity.

"This is expected to be a stressful role, and you'll enter the thick of it pretty much immediately," remarked the executive.

The meteoric ascent of various AI platforms, which have amassed tens of millions of subscribers worldwide, demonstrates the pressing need for such governance guidelines.