As artificial intelligence (AI) technology spreads across industries, ensuring AI safety is emerging as a key national strategic challenge.

The Artificial Intelligence Security Institute (AISI), a subsidiary of the Electronics and Telecommunications Research Institute (ETRI), plans to hold the 'AI Safety Consortium Joint Meeting' at the Grand InterContinental Seoul Parnas in Seoul on October 1st, and to play a leading role in domestic AI safety discussions.

This meeting will share AISI's key achievements, establish the operational structure of the AI Safety Consortium, and discuss in-depth agendas for each policy, evaluation, and research division.

Since its opening in November of last year, AISI has been working to ensure AI safety both domestically and internationally, including conducting safety assessments of independent AI foundation models and international joint evaluations.

The AI Safety Consortium is a collaborative body comprised of 24 companies and organizations, centered around AISI. It contributes to the establishment of government AI governance by proposing policies, developing evaluation methods, and disseminating research results.

In particular, specialized discussions are actively taking place across various sectors, including responding to the European Union's AI regulations, sharing global evaluation cases, and linking domestic research.

The United States and Japan also operate private-sector-led AI safety consortia. AISI is also participating in the establishment of international evaluation standards through collaboration with these organizations. Going forward, AISI plans to expand the consortium's activities by recruiting additional members and operating an executive committee.

ETRI Artificial Intelligence Safety Research Center Director Kim Myeong-ju stated, “AI safety is the foundation of technological innovation and the core of global competitiveness,” and added, “AISI will play a central role in supporting domestic industries and strengthening its role as a focal point for forming AI safety discourse.”

This achievement was achieved through the Ministry of Science and ICT's "Establishing a Foundation for Generative AI Safety Evaluation," and various policy, evaluation, and research activities are expected to continue in the second half of this year.