The simplest design Anthropic could come up with: Have a haiku model quickly judge if a command is harmless or potentially interesting, route all the interestin

The simplest design Anthropic could come up with: Have a haiku model quickly judge if a command is harmless or potentially interesting, route all the interesting commands to a sonnet agent judging it properly. Running two full llms sounds slightly insane, but it's apparently the best we can do (and is reasonably cheap since they don't need a lot of context)