> a corporation can neither "act" nor "think" independently, but rather does so via the humans involved. (Perhaps this would change with corporate-owned AI?) T

> a corporation can neither "act" nor "think" independently, but rather does so via the humans involved. (Perhaps this would change with corporate-owned AI?) This is ignoring that levels of complexity creates new emergent behavior. If you're willing to believe that "AI" could make a corporation think independently, then how is a pile of paperwork running on a substrate of human wetware not the same dynamic? > it seems that prosecutors tend to target corporations simply because it's easier than doing their job properly and pinning down who specifically bears responsibility No, the problem is exactly the sorting through the emergent complexity of the corporation to correctly assign blame. The low-level person who did the actual illegal action is likely sympathetic and mostly judgement proof, and was likely incentivized to break the law by corporate policies. Meanwhile the corporate policies are phrased in terms of abstract metrics that aren't illegal per se, especially how they're written down. Taking the fundamentalist view, that the individual would-be-fall-guy humans should take a hard line and refuse to break the law, doesn't solve the problem - it only increases the level of incentive required until someone is willing to do it. And focusing blame this way helps the higher up management escape accountability since they didn't actually break the law themselves. One correct answer would be to charge all of the involved parties like the criminal conspiracy it is, but the capital-wielding upper classes escaping accountability is a dynamic as old as time.