May 29, 2024
China Proposes Stricter Regulations for Training Generative AI Models

China Proposes Stricter Regulations for Training Generative AI Models

China has introduced draft security regulations aimed at companies providing generative artificial intelligence (AI) services, with a focus on controlling data sources used for AI model training. These regulations were unveiled on October 11 by the National Information Security Standardization Committee, composed of representatives from the Cyberspace Administration of China (CAC), the Ministry of Industry and Information Technology, and law enforcement agencies.

Generative AI, exemplified by systems like OpenAI’s ChatGPT, gains the ability to perform tasks by analyzing historical data and generating new content, such as text and images, based on this training. The committee’s recommendations include conducting security evaluations on the data used to train publicly accessible generative AI models. Content exceeding “5% in the form of unlawful and detrimental information” is to be blacklisted. This category encompasses content promoting terrorism, violence, subversion of the socialist system, harm to the country’s reputation, and actions undermining national cohesion and societal stability. The draft regulations also stress that data subject to censorship on the Chinese internet should not be used as training material for these models.

This development follows the recent permission granted to several Chinese tech companies, including Baidu, to launch their generative AI-powered chatbots for the general public. The Cyberspace Administration of China (CAC) has consistently emphasized the need for security evaluations by companies before introducing generative AI services to the public in April. In July, the regulator released a set of guidelines for governing these services, which were seen as less stringent than the initial draft proposed in April.

The draft security requirements now mandate that organizations involved in training these AI models obtain explicit consent from individuals whose personal data, including biometric information, is used for training. The guidelines also provide detailed instructions on preventing intellectual property infringements.

Around the world, countries are grappling with the development of regulatory frameworks for AI technology. China, aiming to compete with the United States in the AI field, has set ambitious goals to become a global leader in AI by 2030.

Image by Freepik

Related posts

U.K. Government Urged for Stronger AI Oversight

Kevin Wilson

Google’s Gemini AI Gets Affordable: A Boost for Developers

Harper Hall

Microsoft’s AoT vs. CoT: A Leap in In-Context Learning for AI

Harper Hall

This website uses cookies to improve your experience. We'll assume you're ok with this, but you can opt-out if you wish. Accept Read More

Please enter CoinGecko Free Api Key to get this plugin works.