China Tightens Grip on AI Emotional Interactions

China's cyber regulator has proposed draft rules to enhance oversight on AI services that simulate human personalities, aiming to ensure safety and ethical standards. The rules require providers to monitor user emotions and intervene against excessive use, while preventing content that threatens security or spreads misinformation.

China's cyber watchdog has released draft regulations aimed at increasing oversight on AI technologies that mimic human personalities and engage users emotionally. This move highlights Beijing's strategic efforts to steer the swift evolution of AI technologies towards safe and ethical practices.

The proposed regulations target AI products that exhibit human-like personality traits and interact emotionally through various media. Providers will be required to caution users against overuse and step in if users show signs of addiction, signaling a proactive regulatory framework.

Under these proposals, providers must uphold safety throughout the product lifecycle and build robust systems for algorithm checks, data security, and personal information safeguarding. The draft also demands that service providers assess user emotional states and intervene if extreme emotions or addiction are detected, with the content restricted to avoid threats to national security or dissemination of harmful material.

(With inputs from agencies.)