Speaking of prescient, @harris just pointed me to this excellent article: https://estsjournal.org/index.php/ests/article/view/260 This posits the notion of the "moral crumple zone": using humans to take the blame when a complex system like #AI does something wrong. The "prescient" is because it dates back to 2019, early in the AI age. The problem is that we're trying to find humans to take the blame for collective failures. We define companies as "legal persons" financially; I think we need to get much more serious about doing that for civil and criminal penalties. IMO the fault should heavily lie with *owners* (including the stock market) pressuring companies to be careless in the name of speed. If we want to fix that, we need laws to *routinely and quickly* assign fines of millions-to-billions of dollars for every single infraction -- enough to hurt. Don't search for a human to blame -- look to fix the problem. I think that can only be done if the financial risk is greater than the profit.