The problem I have with it is this part of your statement: > And it's not like we couldn't stop them: who was flying that giant plane? Who were those guys rele

The problem I have with it is this part of your statement: > And it's not like we couldn't stop them: who was flying that giant plane? Who were those guys releasing these swarms from the van? Where did they buy these drones? Where did they get the explosives? Just like many other threats, we'd track them down. I think that might not be true. Sure, the slaughterbots video showed a swarm released from a van with operators present. Seems like somebody might notice. But it's easy to imagine a scenario where the operators leave the drones in some out-of-the way place, inactive, without arousing suspicion, and they activate on a timer to carry out their mission. How would you track down who committed the act? A bomb can go off on a delay, but requires the killer to go to the target site to place it. A gun can kill from a great distance, but only with line of sight, the killer needs to be involved at the time of the shot, and even with a silencer it makes a fairly loud noise, drawing attention from those near the shooter. The fictional-for-now slaughterbots scenario would enable killing people at an event without ever going to the event site, without even being in the same city at the time of the event, and at low cost. Remember the scramble to identify and track down the Boston marathon bombers? They attended the race, set down bags with bombs in them, and then left. Fortunately they were both caught on camera and positively identified by surviving victims. What if they didn't have to go to the race at all, but instead went a week before to an abandoned building, inactive construction site, etc and set the hypothetical slaughterbots in place, set to launch later? Would they ever be caught? The combination of low cost, low skill needed by the deployment crew, and low risk of being caught (with corresponding deniability by state actors) is the frightening thing. The other end of it is that you could catch somebody if you had near universal surveillance -- you trace the drones back to their launch site based on reports/footage of them flying midair, and you pull footage from cameras that were watching there too. Or you stop it before it happens with massive data collection online, trying to detect the planning, discussion, or purchasing of equipment. Neither is going to be 100% effective but they'll be enough to get bills passed authorizing more surveillance after a couple attacks. So you could have a situation where terror attacks happen more frequently, or a situation where everybody is subject to a surveillance state granted even more power than the one we have today, or a little from both columns. Not thrilled about that. Obviously the tech is not there yet for the bots to execute attacks with autonomy like depicted in the video. But it seems very possible within this century.