What are the chances that the robot was being controlled by a human and not AI? There have been a lot of "demos" of AI robots that it turns out is a guy in another room controlling it. If I had to bet money, that is what was happening here as well.

This shows the real issue with AI. It isn't in the AI itself, but the people using it. I don't have an issue with them using AI to try and identify the suspect when they are having a hard time doing so. But you can't just take what AI says and assume it is correct and act on it. AI is a tool and needs to be used as such. When it gives you something, it is up to the people to actually check out what it says and follow up on it. People are using AI as if it is the answer but it is just a tool to try and get to the real answer and that is the part that people are ignoring.

I was actually in college in the 1990s, but yes, a middle schooler today with python on a raspberry pi and a pretty simple GPS module could do this.

I didn't say it wasn't abhorrent or alarming. I'm presenting the scenario that this task of "defend this three dimensional coordinate box" doesn't require AI.

Yes, it did. The beacon signals weren't that good back then, neither were the sensors. I had the same problem in the fake robot battles I was involved in.

The answer turned out to be a solution not from Defense industries, but from Genie Garage Door Openers.

The robot doesn't care. The robot's job isn't foreign policy. The robot's job is "here's a box defined by this coordinate cloud, defend it"

Like I said, I programmed it for a fighting robot back in the 1990s. It ain't that complex, and with today's drone factory ships, the Navy can now output this level of AI in killbots at a rate of 10,000 a day.

Kill decisions are simple in comparison: Stay within your predefined geofence, kill anything that moves that isn't transmitting Friend beacon. We don't need AI for that, I coded a form of it in both Basic and Forth back in the 1990s.

And if they don't, some other startup will.

You are putting WAY to much thought into this. Yes, it was not a fancy double blind study, but there are 4 separate files and 4 things they could be assigned to. Anyone with 2 brain cells would listen and assign what they though was the best quality to the original. Then to the one run through the audio cable. Then the banana followed by mud. They fact that at minimum the mud and banana didn't stand out says what you need to know about the quality.

Or time literacy. Permanent!=3 years

Don't be so sure that Republicans aren't Communists and Democrats aren't Fascists.

And apparently "Faster than ever before" is limited to a data set less than 4 centuries old out of 4.5 billion. It's science reporting like this that makes people distrust science.

And the total number of layoffs of American tech workers in 2022-2025 was in the hundreds of thousands.

Far fewer after October 1 when you basically needed to pay an extra full year's salary to hire an H-1b over a citizen.