Theregister2 days agoArtificial intelligenceNo major AI model is safe, but some are safer than othersAnthropic excels in AI safety with Claude 3.5 Sonnet, showcasing lower harmful output compared to competitors. [ more ]