An AI-powered dystopian future seems increasingly inevitable to many these days, but for some, it’s already here.
It’s weird that it’s even legal because the entire premise of it is probability. It’s fundamentally unaligned with ‘beyond reasonable doubt’ because there is literally no capability of doubt in algorithmic processing—quite the opposite, actually; confidence scores.
Hell, I could be wanted for murder in Florida right now. If I ever visit, I’ll find out then in the exact same way as the victims in the article did…
“I was literally on the other side of the planet…”
“Computers don’t lie, kid.”
I have said this before, humans are lazy, inventing new tech needs to assist human minds not replacing them.
In this case the police are taking the word of AI as if it is absolute 120% truth, they are not just using it as a filter to get closer to the guy they want.
I am an IT guy, and I often need to filter long lists of users, devices and apps, I have tried using AI, but I have almost always noted that the result is lacking in some way, and even in those few cases that it doesn’t I just don’t trust the result as I don’t actually see how the filter was applied. I don’t know if the AI model understood what I was asking, or if it made some other random decision.
It is always so weird to me that some guy can be arrested and put in jail for months when there is extremely clear evidence that they should never have been brought in at all.
The procedure should be:
- Computer gives a result
- police investigates it, is it resonable?
- Arrest with a maximum holding time of 48 hours before either release or charges filed
- release pending trial, if not deemed a flight risk