Vietnam will have a national standard for artificial intelligence
The Ministry of Information and Communications has just constructed and solicited comments on the draft of National Standard on Artificial Intelligence.
National standard for artificial intelligence that helps establish the quality assurance and transparency lifecycle of AI modules
National standard for artificial intelligence that helps establish the quality assurance and transparency lifecycle of AI modules
Artificial intelligence (AI) technology is currently developing at an exponential rate, making many important contributions to production, business, services and human life. However, AI is a very complex field and also creates many very worrying challenges. The right understanding of AI to properly and timely grasp the opportunities and challenges from AI is essential for each person, each organization and each country.
Research shows that, at present, international and regional standardization organizations around the world have published technical standards related to AI. However, in the field of AI standardization in Vietnam so far, there is almost no space, there is no specific standard. Currently, Vietnam has only a few standards related to AI.
Facing this situation, the Ministry of Information and Communications has recently developed and received comments on the draft of National Standard on Artificial Intelligence (AI) - life cycle process and quality requirements including 2 part (supermodel of quality and sustainability). The draft standard is proposed by the Ministry of Information and Communications, appraised by the Directorate for Standards, Metrology and Quality, and will be published by the Ministry of Science and Technology.
Standards and specifications can contribute to a secure, high-quality, reliable, and interpretable AI: They form the basis of technical sovereignty, promote transparency, and provide decision-making.
The purpose of the standard is to establish the quality assurance and transparency lifecycle of AI modules. Important quality criteria are identified and specific AI problems will be addressed. To achieve this, the standard presents a set of quality requirements structured in an AI-specific quality metamodel.
It is important to note that not all AI modules impose the same quality requirements. This standard therefore proposes distinctions between AI modules with regard to their safety, security, privacy, and ethical relevance. The safety, security, privacy, or ethics of an AI module require that all quality requirements be reviewed and met, while these requirements are less stringent, when no valid level is given…
This standard applies to all lifecycle stages of an AI module from concept, development, deployment, operation and decommissioning; also refer to different life cycle processes.