A Draft Law In China Proposes That All New AI Products Must Undergo A “Security Assessment.”

On Tuesday, China’s internet regulator revealed a new draft law. The law mandates a “security assessment” for all upcoming AI products before public release. The Cyberspace Administration of China released a draft law stating that national internet regulatory departments must conduct a security assessment for generative AI products. These assessments must be carried out before such AI provides its services to the public.

The Administrative Measures for Generative Artificial Intelligence Services

The draft law aims to guarantee the healthy and standardized development of generative AI technology. It requires that AI-generated content reflects core socialist values and avoids subversive content on state power.

The draft law prohibits AI-generated content that promotes ethnic hatred, extremist or terrorist propaganda, or content that may disrupt social and economic order. The Cyberspace Administration of China is soliciting public intake on the proposed regulations, which will likely become law under Beijing’s centralized political system.

The new regulations have been introduced amidst several flurries of activities. The November launch of San Francisco-based OpenAI’s ChatGPT, which sparked a gold rush in the market, opened the doorway to AI innovations. Now, Chinese companies, such as Alibaba, NetEase, JD.com, and ByteDance (TikTok’s parent company), are rapidly developing services that mimic human speech. This rush to develop the next innovative AI technology has prompted regulation concerns.

#BREAKING China to require 'security assessment' for all new AI products: draft law pic.twitter.com/Hc5YumYywR

— AFP News Agency (@AFP) April 11, 2023

Increased Concerns About AI Advancement

The rapid progress of AI has raised concerns about its potential for disinformation and misuse, such as deepfake images and manipulated videos. In response, Beijing implemented new rules in January that mandate businesses offering deepfake services to verify the real identities of their clients. The law also requires these businesses to label deepfake content to prevent confusion appropriately.

The Chinese government has cautioned that deepfakes threaten social stability and national security. Despite ChatGPT being unavailable in China, Chinese users are using VPNs to access the American software and leveraging it for essay writing and exam preparation.

China aims to lead the global AI industry by 2030, with the sector predicted to add $600 billion annually to the country’s GDP. Despite this, domestic efforts to develop competitive AI products have struggled due to US restrictions on chip imports and Beijing’s strict censorship.

Featured image from economictimes.com