Anthropic admits AI 'not perfectly aligned' with human values

(theguardian.com)

9 points | by fittingopposite 5 hours ago ago

6 comments