Robot dogs armed with AI-aimed rifles undergo US Marines Special Ops evaluation
Robot dogs armed with AI-aimed rifles undergo US Marines Special Ops evaluation
This definitely can’t go badly.
Can it?
No.
Well, yes it can go bad. I think they forgot the self-replication mechanisms.
What? Humans as a species suck. all hail the AI overlords.
I was involved in discussions 20-some years ago when we were first exploring the idea of autonomous and semiautonomous weapons systems. The question that really brought it home to me was “When an autonomous weapon targets a school and kills 50 kids, who gets charged with the war crime? The soldier who sent the weapon in, the commander who was responsible for the op, the company who wrote the software, or the programmer who actually coded it up?” That really felt like a grounding question.
As we now know, the actual answer is “Nobody.”
It’s all based on geography.
If the school is located in a mineral rich area or underground oil field, then that wasn’t a school, it was a military base and those weren’t students they were terrorists.
If the school is located in aa area that lacks any natural wealth, then the robots have become autonomous and acted without control by anyone. It was an accident.
The issue people are worried about is that no one is making the decision to kill kids, it’s the AI making the call. It’s being given another objective and in the process of carrying that out makes the call to kill kids as part of that objective.
For example, you give an AI drone instructions to fly over an area to identify and drop bombs on military installations, and the AI misidentifies a school as a military base and bombs it. Or you send a dog bot in to patrol an area for intruders, and it misidentifies kids playing out in the streets as armed insurgents.
In a situation where it’s human pilots, soldiers, and analysts and such making the call, we would (or at least should) expect the people involved to face some sort of repercussions- jail time, fines, demotions, etc.
None of which you can really do for a drone.
And that’s of course before you get into the really crazy sci Fi dystopia stuff, where you send a team of robots into a city with general instructions to clear it of insurgents, and the AI comes to the conclusion somehow that the fastest and most efficient way to accomplish that is to just kill every person in the city since it can’t be absolutely sure who is and isn’t a terrorist
That’s also a legal issue with autonomous driving cars.
Autonomous cars can also get into basically the trolley problem. If an accident is unavoidable, but you can swerve and kill your own passenger to avoid killing more people in a larger wreck, should you? And would that end up as more liability for whoever takes the blame?
Are we talking truly autonomous vehicles with no driver, or today’s “self-driving-but-keep-your-hands-on-the-wheel” type cars?
In the case of the former, it should be absolutely the fault of the manufacturer.
Say there is a car with no human driver, that is being sold as requiring “no human input other than set destination, stop, and go”.
If that vehicle crashes, you think the person who bought the car (the passenger) has legal liability, and not the manufacturer?
That’s like being a passenger on a bus and getting sued if the bus driver hits a parked car.
When an autonomous weapon targets a school and kills 50 kids, who gets charged with the war crime?
When a human in a plane drops a bomb on a school full of kids, we don’t charge anyone with a war crime. Why would we start charging people with war crimes when we make the plane pilotless?
The autonomy of these killer toys is always overstated. As front-line trigger pullers, they’re great. But they still need an enormous support staff and deployment team and IT support. If you want to blame someone for releasing a killer robot into a crowd of civilians, its not like you have a shortage of people to indict. No different than trying to figure out who takes the blame for throwing a grenade into a movie theater. Everyone from the mission commander down to the guy who drops a Kill marker on the digital map has the potential for indictment.
But nobody is going to be indicted in a mission where the goal was to blow up a school full of children, because why would you do that? The whole point was to murder those kids.
Israelis already have an AI-powered target-to-kill system, after all.
But in 2021, the Jerusalem Post reported an intelligence official saying Israel had just won its first “AI war” – an earlier conflict with Hamas – using a number of machine learning systems to sift through data and produce targets. In the same year a book called The Human–Machine Team, which outlined a vision of AI-powered warfare, was published under a pseudonym by an author recently revealed to be the head of a key Israeli clandestine intelligence unit.
Last year, another +972 report said Israel also uses an AI system called Habsora to identify potential militant buildings and facilities to bomb. According the report, Habsora generates targets “almost automatically”, and one former intelligence officer described it as “a mass assassination factory”.
The recent +972 report also claims a third system, called Where’s Daddy?, monitors targets identified by Lavender and alerts the military when they return home, often to their family.
Literally the entire point of this system is to kill whole families.
As we now know, the actual answer is “Nobody.” the 50 kids who gets designated as “terrorists” afterwards.
FTFY - it’s the American way.
Get the Super Soakers ready, and fill them with saltwater!
Electronics really don’t like saltwater…

Totally unrelated video about normal robot dogs without guns on them