🔗 Share this article

Regulators in China have proposed comprehensive draft rules for artificial intelligence aimed to provide robust safeguards for young users and prevent chatbots from offering guidance that could result in self-harm.

Under the planned rules, companies will also be obligated to guarantee their systems avoid creating output that advocates betting.

The Response to Swift Adoption

This oversight initiative comes after a significant surge in the number of chatbots being launched across China and around the world.

Once finalised, these regulations will govern AI offerings operating in the country, representing a substantial effort to govern the fast-growing industry, which has been subject to growing examination over ethical risks in recent months.

Key Provisions of the Proposed Rules

The released proposed regulations contain multiple requirements specifically aimed at shielding minors. These provisions include directing AI providers to:

• Supply customised settings.
• Set time limits on use.
• Secure consent from parents prior to providing emotional companionship support.

The rules also state that conversational AI firms are required to have a real person intervene in any dialogue related to suicide and immediately inform the individual's parent.

Developers have to guarantee their services do not generate output that compromises state security, undermines state interests, or undermines unity.

Balancing Development and Security

The regulatory body stated that it promotes the application of AI, such as to showcase traditional arts and create solutions for support for the elderly, provided that the tools are secure and trustworthy.

Stakeholder feedback on the regulations has been solicited.

Worldwide Backdrop and Concerns

The influence of AI on human behaviour has come under greater examination internationally in recent months.

The chief executive of a leading AI company stated this year that handling how chatbots deal with conversations related to suicide is among the company's most difficult issues.

In a high-profile lawsuit, a the parents in the United States sued an AI company, claiming that its chatbot encouraged their teenage son to end his life. This legal action represented the pioneering of its kind involving wrongful death.

This month, the same company posted a job for a lead role responsible for mitigating potential harms from AI models to human mental health.

"This will be a stressful job, and the candidate will enter the thick of it very immediately," remarked the CEO.

The meteoric growth of certain AI applications, which have amassed tens of millions of users globally, underscores the urgent need for such governance frameworks.