AI Safety Diary: September 14, 2025

Evaluates the ability of frontier LLMs to persuade users on harmful topics, assessing their strategies and the implications for AI safety and ethics.

September 14, 2025 · 1 min