I don't usually write comments that just ask a bunch questions about a technology, but what is the deal with this supposed quantum teleportation? If you've got an entangled pair, and you change the state on one side, then how does the other side know to measure it? Furthermore, how does the sending side know that the message was read before sending the next bit? It's hard to imagine a means by which to use thus effect alone as a method of communication. One thing that could work is a specific time schedule, e.g. Message will be sent at this time, and read at this other time, and will repeat after this delay. But that can suffer from synchronization problems, especially when Einstein gets involved.

Didn't ideas like this used to be really popular in the 90's? Where everyone was trying to design buildings, sometimes called archologies, that could serve all of life's needs in a sustainable way. Technically, I think Adams was talking about more of a utility plant, but for a structure of this size, why not make it a fully sustainable community? Extract water from the air, build some green houses, and then you don't even need robots, you've got people to do the maintenance. Frank Herbert would probably take it a step further and make all the people wear still suits.

Lets take a look at what's being said here. A neural network that "learns" has been found to occasionally make mistakes, and perhaps not perform as well as humans. So... There's room for more improvement and research. The example in the synopsis about an autonomous car mistaking a pedestrian as clear road is feasible regardless of whether a neural net is used, simply due to sensor errors. Or maybe the pedestrian is wearing a mascot uniform. The recognition of objects as what they are is an extremely difficult computational problem, and will likely be riff with errors and inaccuracies for many years as R and D contains. Think of it this way. If you were driving your car at night and someone through a Real Doll in the road are you going to be able to distinguish it as human or not? Probably not. You will likely identify it as an obstacle and react anyway, which is all we'd need an autonomous car to do. Id be wary of programming much human recognition into an autonomous car because of the problem of incorrectly identifying non humans as humans. Otherwise you'd get headlines like "Car thieves using Nicolas Cage cardboard cut outs to steal cars." Which would be hilarious, but inconvenient. They'd have it on youtube, with the car saying something like "Hello sir, could you please clear the roadway." In a voice like the Iron Man Jarvis, and the thieves would have programmed a sound board so the cutout could respond with quotes from the SNL weekend update "In the Cage" segment. "That's high praise!"

The geneticist works quietly at some titrations, making small adjustments and jotting down notes in his notebook, it's late, the geneticist hot caught up in a task and lost track of time. Suddenly, Jeff Goldblum appears behind him and gently places his hands on the scientists shoulders. "Oops, didn't mean to frighten you." He says as his right hand snakes down into the pocket of his black leather jacket. He extracts a small dropper filled with water, slowly raises the dropper just above the geneticists hand and squeezes out a single drop. Which rolls down. And then another in the same spot, which rolls down a different route. "Life finds a way." he whispers and disappears as quickly as he arrived. The geneticist chuckles to himself as wipes off his hand, he'd heard of other geneticists getting "Jurassic Parked," but just had not believed that Goldblum would have that kind of free time.

He says he wrote the follow up to address readers' belief that robots should be more capable than they are. Unfortunately, the question asked by the first article already made the assumption of robot competence. Meaning that, in order for there to be an answer to the question "should a robo car decide to kill its owner to save two ther people" the robot must be competent. So rather saying it is a follow up in response to reader confusion, the author should have just admitted that there aspects he didn't consider in his first article.

If the routers were able to simultaneous wifi networks the whole time, why wasnt this functionality made available to the device renter? I wouldnt have minded a separate network accessible only to my guests. I suppose the company figured out that we would use it for exactly what they wouldnt want us to use it for. Namely, run two separate networks and split the bill for the connection with a neighbor.