PwC Consulting said on the 21st that it will hold an emergency expert roundtable with Yu Dong-soo of the Democratic Party of Korea, a member of the National Assembly's National Policy Committee, on the "Mitos incident."
The emergency expert roundtable will be held at 10 a.m. on the 23rd at the Sapphire Hall on the second floor of the FKI Tower Conference Center in Yeouido, Seoul, under the theme of the decision to withhold the public release of the frontier artificial intelligence (AI) Mitos and cyber crisis response strategies for the nation and corporations.
The latest AI model "Claude Mitos Preview (Mitos)," developed by U.S. AI corporation Anthropic, is said to autonomously detect security vulnerabilities across global operating systems, web browsers, and networks, and in some cases even derive attack paths, sending shock waves through the global security industry.
Anthropic judged that if such a frontier AI model, with advanced autonomy and a broad range of applications, were released to the public, it could have a significant impact on national critical infrastructure, financial systems, and overall public safety. It therefore decided, for the first time ever, not to release the model to the public.
The roundtable was organized to share the context of changes brought by the emergence of frontier AI technology with government, public sector, industry, academia, and security experts, as the "Mitos incident" has become a global issue. In particular, in an environment where AI-based cyber threats are becoming a reality, it will examine preemptive response tasks at the policy, regulatory, and industrial levels, focusing on what standards and perspectives the nation and corporations should adopt.
The roundtable will broadly discuss: ▲ the spread of AI-based cyber threats ▲ vulnerabilities across the nation and industry ▲ directions for responses at the policy, regulatory, and industrial levels ▲ governance issues for an era when AI is not publicly released.
Korea University Professor Lee Sang-geun will deliver the keynote, outlining the technological and security shifts driven by frontier AI.
In the following panel discussion, Choi Jang-hyeok, Samil PwC AI Trust chairperson (former vice chairperson of the Personal Information Protection Commission), will serve as moderator.
Regarding the roundtable, Chairperson Choi noted, "In an era when AI is too powerful to be made public, this will be the first official forum to ask how organizations such as the nation and corporations will survive."
Panelists include Im Jong-in, adviser at Kim & Chang; Lee Seong-yeop, professor at Korea University Graduate School of Management of Technology; Ko Nak-jun, director general of prevention and coordination at the Personal Information Protection Commission; Lim Jeong-gyu, director general for networks at the Ministry of Science and ICT; Yu Yeong-jun, director of digital finance policy at the Financial Services Commission; Oh Jung-hyo, head of AI and strategy at the Financial Security Institute; Lee Sang-geun, head of the AI Security Research Institute at Korea University; Cheon Jeong-hee, professor of mathematics at Seoul National University (Cryptolab CEO); Song Yeong-sin, managing director (CISO) at Shinhan Bank; Hong Gwan-hee, executive vice president (CISO) at LG Uplus; Park Hyun-chul, executive director at PwC Consulting; and Kim Dae-hwan, CEO of Somansa, among other security experts from government, academia, and industry.
Park Hyun-chul, leader of Risk & Cyber Services at PwC Consulting, said, "As we enter a stage where whether AI technology is made public itself must be treated as a risk, simple discussions about using the technology are no longer sufficient," adding, "I hope this roundtable becomes a starting point for organizing the response standards required across policy, regulation, and industry, and for deriving future direction."
Registration is available on the PwC Consulting website.