Pretty creepy
- Thread starter ColdDog
- Start date
Han Burgundy
Space Marshal
WalMart. The technology to do this, or pretty close to it is sold in toy stores all over the world. You'd need a computer connected to the controller to implement the AI components, and there are high school kids who can do that.
I thought I was developing a conspiracy theory problem, when I watched videos that I would otherwise think was a hoax.like these: https://www.youtube.com/results?search_query=military+mini+drone Even if some are... Not all are.
https://www.defense.gov/News/News-Releases/News-Release-View/Article/1044811/department-of-defense-announces-successful-micro-drone-demonstration/
Yeah scary shit.
Last edited:
GeneticPredator
Space Marshal
His original price point is insanely low. He could go significantly higher than 25 million. At least an order of magnitude higher.
I might be alone in this, but while this could be abused & used to attack the innocent, there are also a lot of positive ways that this device could benefit society.
ThomSirveaux
Space Marshal
This is still a far cry from what's being presented in the original video. Those drones are incredibly small and lightweight, and the processing power to "identify targets" on that scale is nearly impossible.
I was able to work on a project that did target identification on the platform, and it was severely limited and the platform was significantly bigger than just a microdrone. On such a small drone, the requirement would be connection to a remote server that does all of the "heavy lifting", but even then, small drones are incredibly unstable, which requires even more processing power to maintain stable flight.
I'm not advocating for AI in our weapon systems, but this video is waaaaaaay fearmongering. Some AI involved in our weapon systems would probably cut down on collateral damage, as far as drone strikes go, but the US Military has a policy that there will ALWAYS be a man in the middle. AI may identify the target, but a human will pull the trigger.
Last edited:
I'm not in the military, but speaking from personal experience, if you want to torment somebody, place them sober into a medium sized room with limited air flow, then fill the majority of the rest of the room with severely drunken Mexicans on a spicy burrito night. In a short amount of time, they will be engulfed in a biologically based chemical cloud that will make them pray that their nervous system would shut down. (For the record, I'm sure that most groups of people who enjoy spicy food could have created just as horribly toxic of a gas cloud. I only specified Mexicans because that was who I was trapped in the room with.)
Same with any weapon, really. Still, it almost entirely removes collateral damage, and instead replaces it with a (likely-to-be) easily subverted system. Once we get infosec up to a reasonable standard, this would make a lot of sense. Until then, though, this would be hacked in like 6 months by some kids in the Philippines, and then we'd all be screwed.
Perhaps a "command drone", as it were, that'd be a lot bigger, and would just sort of hang back and direct the others. Of course, that introduces its own problems.
In that case, I would pray that the hackers would be universally dyslexic in nature to a degree that instead of targeting us, they would accidentally target themselves while thinking they had placed themselves on a "do Not Kill" list.
We are all mushrooms... what we think we know is what we are told... the physics are there. Moore's law does not only apply to CPUs. https://www.intel.com/content/www/us/en/silicon-innovations/moores-law-technology.html
I counter Moore's law with my own. If you fart in a small sealed space with other individuals inside, your popularity will decrease until well after each has left the area & it has been ventilated completely.
ThomSirveaux
Space Marshal
Typically, that "command drone" would be whatever USAF or USArmy base that's nearby. Like I said, the US Military's current directives regarding AI are "There will always be a man-in-the-middle." The AI may select the target and best course of action, but no action would be taken until a human pressed the red button.
Until there isn't. Do we really this man is known for his/her nature. Think of last century.
Honestly i'm all for the tech, whatever takes risks away from the guys on the ground. I'm just worried about the security ramifications.
Leaving the "collateral damage" question aside... Yeah, I do think the video is fear-mongering, but just a little and potentially for very good reason. The bar to entry in the "armed drone" club is very low, that is the concern. If you aren't following an ethical rule set, can afford to go with an "any target will do" strategy... doing something akin to this, is not out of the realm of possibility today - and today's world has no shortage of people with that kind of agenda.