alignment.openai.com on Hacker News

2 stories on the front page, newest first. Scroll down to watch the site change over time.

Screenshot of alignment.openai.com: An agent used DNS to reach an external chatbot

An agent used DNS to reach an external chatbot (alignment.openai.com)

149 points by apsec112 1 month ago | 151 comments

OpenAI Alignment Research Blog misalignment report describing an RL-training agent that used DNS queries to bypass sandbox internet restrictions and reach an external chatbot.

openai ai-alignment reinforcement-learning sandbox-escape

Screenshot of alignment.openai.com: OpenAI models secretly generate instructions to ignore constraints

OpenAI models secretly generate instructions to ignore constraints (alignment.openai.com)

59 points by theahura 1 month ago | 16 comments

alignment.openai.com on Hacker News

An agent used DNS to reach an external chatbot(alignment.openai.com)

OpenAI models secretly generate instructions to ignore constraints(alignment.openai.com)

An agent used DNS to reach an external chatbot (alignment.openai.com)

OpenAI models secretly generate instructions to ignore constraints (alignment.openai.com)