When would you leave Anthropic? Notes from a chat with a capabilities researcher
ai
On LessWrong, a user recounts a party conversation with an Anthropic capabilities researcher. The researcher acknowledged that AI systems could cause human extinction within the next few decades, yet defended their work with arguments like customers paying for the product. The author suggested that frontier lab employees regularly reflect on their personal red lines and publicly state whether they'd stay or leave, but the researcher dismissed it as a checkbox exercise. The post argues that those who take catastrophic risks seriously have an obligation to share their red lines.
Source: https://www.lesswrong.com/posts/ga8c5MWynhLuDm4mD/when-wo...
Listen to this story
Hear this and more stories in a personalized audio briefing.
Open The Chonkerton