Anthropic Expands AI Mythos Access to Samsung, SK

New partners include Samsung Electronics, SK Hynix, SK Telecom and institutions from 15 countries

U.S. artificial intelligence (AI) company Anthropic has expanded the countries and permissions for access to its AI Mythos, which possesses security vulnerability detection capabilities at the level of human experts. Domestic companies reportedly include Samsung Electronics and SK Hynix.

Anthropic announced that 150 institutions from 15 countries have been added as partners to Project Glasswing, a security consortium based on Mythos. While Anthropic unveiled Mythos, which detects security vulnerabilities, in April, it had limited access to only 50 institutions under Project Glasswing due to concerns over misuse.

Anthropic stated that the newly participating institutions cover new industrial sectors such as power, water, healthcare, telecommunications, and hardware, which were lacking among the initial participants. It added that while the 15 countries where newly authorized partners are located include Canada, Australia, France, Germany, Japan, and South Korea, issues at these institutions could affect over 100 million people and cause global security repercussions.

Domestic companies among the newly authorized partners include Samsung Electronics, SK Hynix, and SK Telecom. Samsung Electronics and SK Hynix recently participated as strategic infrastructure partners in Anthropic's Series H funding round.

Anthropic revealed that initial partners, approximately 50 institutions, identified over 10,000 security flaws classified as "high" or "critical" severity through security checks using Mythos.