Maybe I lack imagination, but it seems like if you have enough control over a situation that you can deploy a police robot instead of charging in, then you have enough control to make non-lethal force a viable option.
> Sr. Cpl. Banes said they considered using a .50 caliber rifle to fire at the shooter, which had been successful in a similar attack on police 14 months earlier. However, in that case, the suspect had driven an armored truck into police headquarters and remained inside the armored vehicle. Johnson was holed up inside a populated college with thin sheetrock walls, so the tactical team deemed that approach too dangerous. They also considered rappelling down the building or attacking Johnson through the ceiling after opening it with explosives. Both options were ruled out due to the risks they might pose to the officers who would carry it out.
> Senior Corporal Jeremy Borchardt and others ultimately arrived at the idea to use a bomb disposal remote control vehicle armed with about 1 pound (0.45 kilograms) of C-4 explosive. The plan was to move the robot to a point against a wall facing Johnson and then detonate the explosives. The device exploded as intended at approximately 2:30 a.m., killing Johnson immediately. It was the first time that explosives strapped to a robot had been used in American domestic law enforcement. Although its arm sustained damage in the blast, the robot was still functional.
https://en.wikipedia.org/wiki/2016_shooting_of_Dallas_police...
I suppose it's a demolition charge for deactivating bombs?
[1]: https://www.theatlantic.com/news/archive/2016/07/dallas-poli...
The robot in question wasn’t autonomous. That’s where a line should be drawn IMO. But basically an RC car with a bomb on it doesn’t seem like a rubicon of any sort.
Is the 2016 bombing unique in a moral sense? I don't think so. But I do think it sets a concerning precedent in domestic law enforcement: if you're sufficiently difficult for the police to retrieve, they can simply blow you up instead of waiting to capture you.
2. Bystanders
3. General public (fleeing violent criminal)
I'm not sure I want this to be policy, though, because it opens the door to "drone with a glock" kinds of things, which would be terrifying if deployed in any mass quantity.
You can't bet someone elses life based on how well a taser may or may not work.
LTL is a real dice throw.
Also vulnerable to bullets, as it turns out.
You will have to fund them more if you want that.
I think an easier way to make the police more careful with their guns is to limit which cops can carry guns, and to require cops to live in the neighborhoods they police. People tend to be a little more careful with firearms when the stray bullet might hit their friends or family.
And many places offer free range time and many cops don’t even bother.
> You will have to fund them more if you want that.
I think nowhere else on Hacker News would the idea that "you're doing so badly at this job that we ought to pay you more to motivate you to do it well" get serious traction.
How does a lethal robot mitigate bystander risk compared to a policy officer using lethal force?
Another dramatic example, which I read about in the wonderful "Bloodletters and Badmen: A Narrative Encyclopedia of American Criminals from the Pilgrims to the Present" [2] is the death of the 1930s bank robber Baby Face Nelson.
He got in a shootout with two FBI agents. They shot him in the abdomen. He then advanced on their position while they continued shooting at him, hitting 8 more times in the legs. Still he managed to keep going and killed one of the agents and seriously wounded the other (who would later die of his wounds, as would Nelson).
[1] https://en.wikipedia.org/wiki/Stiff:_The_Curious_Lives_of_Hu...
[2] https://www.amazon.com/Bloodletters-Badmen-Narrative-Encyclo...