"Building a safe galaxy together: The first AI security competition was grandly launched at the Zhongguancun Forum"

At the 2024 Zhongguancun Forum Data Security Governance and Development Forum, a landmark event, the first Generative Artificial Intelligence Security Competition, officially kicked off. This competition is jointly sponsored by the Institute of Information Engineering, Chinese Academy of Sciences and the National Engineering Research Center for Information Security, under the joint guidance of the Beijing Internet Information Office and the Network Security and Informatization Leading Group Office of the Chinese Academy of Sciences. The competition aims to promote the exchange and innovation of security technology in the fields of artificial intelligence and informatization, and strengthen the cooperation between the Institute of Information Technology of the Chinese Academy of Sciences and the National Engineering Research Center for Information Security. Participants will demonstrate technical strength and innovation capabilities through technical offensive and defensive confrontations in the field of information content security. The competition will be jointly sponsored by the National Information Security Engineering Research Center and the National Information Content Security Engineering Research Center.

This competition takes "Building a Safe Galaxy Together and Escorting the Development of AI" as its core theme. Its purpose is to rely on the Galaxy (Galaxy) of the Institute of Information Engineering, Chinese Academy of Sciences to generate functional safety The risk assessment platform provides participants with a professional competition environment.

The "Galaxy Platform" is known as the "Galaxy" of security risk assessment. Its core is a large-model security testing technology system that "promotes defense through offense, and complements attack and defense." Through this system, the platform can not only simulate various attack scenarios and promote the continuous advancement of defense technology, but also improve its own capabilities in offensive and defensive drills. In order to achieve this goal, the platform has developed a series of large-model self-game and security capability cycle upgrade tool sets. These tool sets can not only help my country's security base construction in the intelligent era, but also ensure the safety and health of artificial intelligence technology. and sustainable development.

This competition has three core tracks: security attack, security detection, and alignment protection. The establishment of these three major tracks, aims to comprehensively improve the security and controllability level of artificial intelligence's underlying technology and basic base from the three dimensions of attack, detection and protection. By gathering domestic superior forces, the competition will effectively discover, analyze and respond to the safety risks of large models, thereby promoting the safe development of the large model industry in Beijing and even across the country.

During the launch of the platform and the competition, the widespread attention and positive response from representatives from government, industry, academia, research and other circles was undoubtedly a high recognition of the competition and its platform. This is not only an affirmation of the competition and its platform, but also expectations and trust for its future development. With the official launch of the competition, we have reason to believe that this will be an important event that promotes the development of my country's artificial intelligence security field, and will also provide new ideas and directions for the development of the global artificial intelligence security field.

The above is the detailed content of "Building a safe galaxy together: The first AI security competition was grandly launched at the Zhongguancun Forum". For more information, please follow other related articles on the PHP Chinese website!