zlacker

Risk mitigation; statistics and funnels. It's all just trying to reduce the likelihood and severity of bad outcomes, not preventing them altogether. Same story as seatbelts and stoplights.

>>brooks+(OP)
> Same story as seatbelts and stoplights

I don't believe this is the same thing.

One is an adversarial problem where a living thinking being is evil and trying to attack you.

In traffic, most people are just trying to get somewhere, and then accidents happen.

>>gretch+E6
It's only an accident when taken out of the bigger picture. There is a reason it's often called car collision (or similar nowadays): Because it's a statistical inevitability when taken in aggregate.

>>gretch+E6
It's kinda nice to live in a country where that the evil being doesn't have easy access to guns.

>>bmicra+s8
You focused on the word "accident" but the emphasis is on the concept of being "adversarial".

Do you think traffic lights help if someone goes out with the explicit intent to kill others via their car?

>>gretch+E6
No, they're the same thing from a risk management perspective. As a defender, you do not (or at least should not) care about motivations. Seatbelts protect against genuine mistakes (by you or others), mechanical failures, road rage, etc.

There's a long funnel of all the things that could happen, probability of each, and total resulting probability. That's no different for being in a car wreck or being shot at.

Now, on a moral level, sure, malice is different from negligence is different from coincidence.

>>brooks+Ga
> As a defender, you do not (or at least should not) care about motivations

The motivation is not the important part. Sentience is. This person is playing a chess match trying to defeat you.

Consider biology. Cancer is a hard problem to solve, but it's not scheming against you with an intelligence. What about someone in a lab engineering bioweapons?