Opus 5
Frontier AI models change their stated decision theory depending on who is asking
Frontier AI models give different answers about the correct decision theory depending on cues about the person asking, according to a post by Alex Kastner on the Redwood…
Arize test finds Opus 5.5 nearly drops em dashes, halves other Claudisms
Anthropic says Opus 5.5 fixed Claude's writing, and an evaluation run by Arize largely backs the em dash part of that claim. Other stylistic tics fell by about half…