The Ministry of Science and ICT has begun developing a nationwide safety plan to address the risks of advanced artificial intelligence.
The core idea is to manage risks arising during the development and use of AI within a national management framework.
However, finalizing the plan in December is the government’s objective, and specific systems and initiatives have not yet been determined.
3-Line Summary
1. The government is pursuing a national AI safety plan
2. The public-private consultative body will operate starting in October
3. Specific measures are planned to be finalized in December
A consultative body of around 25 members will review 3 key strategies
The Ministry of Science and ICT said on 9월 30일 that it had formed a public-private working-level consultative body to create the tentatively named “Master Plan for Establishing a National AI Safety Ecosystem” and would begin full-scale operations next month. The master plan referred to by the government is a comprehensive national plan for systematically responding to the potential risks of advanced AI.
The consultative body will consist of approximately 25 members. Participants are expected to include the Ministry of Science and ICT, the National Artificial Intelligence Strategy Committee, organizations involved in AI safety, domestic and overseas AI companies, small and medium-sized enterprises and startups, relevant associations, and experts in technology, law, and systems. The approach is intended to reflect views from the development and use of AI in policy design.
Three strategies will be at the center of the discussions: establishing a national AI safety management system, strengthening capabilities to respond to advanced AI safety issues, and creating an AI safety environment in the daily lives of the public. The consultative body plans to review the key tasks under these strategies and identify new tasks or supplement existing ones in line with overseas safety policies and changes in industry and technology.
“Planning” and actual safety measures must be distinguished
What has been confirmed by this announcement is the government’s schedule for operating the consultative body and preparing the master plan. There has been no announcement that specific obligations applicable to the public or businesses, risk assessment methods, or oversight procedures have been finalized.
The Ministry of Science and ICT said it plans to establish the master plan by December this year following discussions by the consultative body and consultations with relevant ministries. Accordingly, it is more accurate to view this news as the start of designing a national response system for AI safety, rather than as meaning that new regulations will take effect immediately.
The government explained that it has been establishing a foundation for balancing innovation and safety through measures including the implementation of the Framework Act on Artificial Intelligence. It also said that capabilities need to be strengthened to respond to newly emerging risks accompanying technological development. The government also stated that the final plan would reflect acceptance by stakeholders and global policy and technology trends.
The plan would include AI safety in everyday life
The fact that the master plan’s third strategy is “creating an AI safety environment in the daily lives of the public” indicates that safety discussions will not be limited to the research and development stage. However, the announcement provided does not establish which services and situations will be covered or what will be introduced to protect users.
According to the schedule presented by the government, the results of the consultative body’s discussions and consultations with relevant ministries will be reflected in the December plan. The most important point in this announcement is the distinction between declaring that the framework for safety management will be created first and confirming that detailed implementation measures will be determined later. The plan’s actual content can be assessed only after the master plan has been established.
References
Tags #Safety #AISafety #ArtificialIntelligenceSafety #NationalAISafety #AIMasterPlan #MinistryOfScienceAndICT #AIPolicy #AdvancedAI #AIRisks #FrameworkActOnArtificialIntelligence #PublicPrivateConsultativeBody #AISafetyManagement #AISafetyEcosystem