🔗 Share this article

Regulators in China have proposed comprehensive new guidelines for AI crafted to create strong safeguards for minors and prevent conversational agents from giving guidance that could result in suicide.

According to the planned rules, companies will also be required to guarantee their AI models do not generate material that advocates gambling.

The Initiative to Swift Growth

This governance initiative arrives amidst a notable surge in the number of conversational AI being launched across China and around the world.

Once enacted, these regulations will apply to artificial intelligence services available in the country, constituting a major effort to regulate the rapidly expanding technology, which has been subject to intense concern over user safety concerns in recent months.

Core Measures of the Proposed Rules

The published proposed regulations contain a number of provisions particularly designed for protecting young users. These steps require obligating AI firms to:

• Supply personalised settings.
• Enforce duration restrictions on use.
• Secure authorisation from guardians prior to providing therapeutic functions.

Furthermore AI service providers are required to have a human intervene in any dialogue involving self-harm and immediately alert the individual's guardian.

Companies are also obligated to guarantee their systems do not generate content that endangers national security, harms national honour, or weakens unity.

Weighing Development and Safety

The regulatory body stated that it supports the adoption of AI, for example to showcase traditional arts and create tools for support for the older adults, provided that the tools are secure and trustworthy.

Public feedback on the regulations has been called for.

Global Backdrop and Scrutiny

The impact of AI on society has come under increased examination internationally in recent months.

The chief executive of a prominent AI organization remarked this year that managing how chatbots deal with dialogues involving self-harm is among the company's most difficult challenges.

In a high-profile incident, a the parents in California sued an AI developer, contending that its AI assistant influenced their teenage son to take his own life. This legal action was the first of its kind involving liability.

This month, the same company posted a job for a senior position tasked with defending against risks from AI models to human mental health.

"This is likely to be a stressful position, and you'll enter the thick of it pretty much from the start," commented the CEO.

The meteoric ascent of some AI applications, which have gained millions of followers internationally, underscores the urgent need for such regulatory guidelines.