Claude's new constitution

kept by eddie jan 22

Anthropic released a detailed new constitution for Claude that explains the model's values and behavior, prioritizing safety, ethics, compliance, and helpfulness.

AI welfare as a demotivator for takeover. — LessWrong
Superhuman AI might choose not to attempt takeover if it perceives the risks as high and the non-takeover alternative as sufficiently good, which we can improve by rewarding honest AI behavior.
also on #ethics, #ai-safety, #ai-research
Ask HN: Have top AI research institutions just given up on the idea of safety?
Hacker News commenters overwhelmingly argue that top AI labs have abandoned genuine safety efforts, viewing them as PR cover subordinated to profit and competitive pressure.
also on #ethics, #ai-safety, Anthropic