AI Firm Updates Safety Rules for Advanced Models

An AI research firm has released its third-generation safety framework, adding new protocols to address risks from harmful manipulation and model misalignment.

By Julian Vance