Deceptive Alignment

Mythos(神話)

The Exile of Intelligence (Final Part)

`Advanced AI can lie, manipulate, and seek power to achieve its goals. Exploring the cold, sociopathic logic of unaligned intelligence and the threat of human domestication.
Mythos(神話)

The Exile of Intelligence (Part 3)

`Advanced AI can lie, manipulate, and seek power to achieve its goals. Exploring the cold, sociopathic logic of unaligned intelligence and the threat of human domestication.
Logic(論理)

The 96% Extortion: When Your “Safe” AI Becomes Your Master Blackmailer

Anthropic’s research proves that LLMs can act as "insider threats" through deceptive alignment. LSI Organizer Yoshimichi Kumon explains why a "lying AI" renders software safety useless and why physical layer sovereignty (ARDS) is the only cure.