I think this comment significantly underestimates how good Opus 4.6 is at exploitation given decent scaffolding and lots of inference compute.
I find that surprising but expect you to know better than I do there. As such you’relikely right, especially if we use your definition of “lots of inference compute”.
I think this comment significantly underestimates how good Opus 4.6 is at exploitation given decent scaffolding and lots of inference compute.
I find that surprising but expect you to know better than I do there. As such you’relikely right, especially if we use your definition of “lots of inference compute”.