So...there's a lot of AI headlines lately and none of them are good. Anthropic apparently told the DoD (now called the DoW) that they wouldn't allow their tech to be used for mass surveilance or unrestricted military use, because they're trying to be more ethical than other AI companies. They were told to loosen their restrictions or get blacklisted, since other companies would allow that.
Let's just say...NONE OF THIS IS OKAY! If we use AI for mass surveillance we might be turning 1984 into a thing, and if we allow unrestricted military use, we could be causing terminator or at least "wargames" to be a thing. Speaking of "wargames", that brings us to a second story. AI apparently really REALLY likes nukes in simulated war games scenarios, using them 95% of the time. So...think about that previous headline and combine it with this one. And imagine what the end result will be. I mean...duh. Wtf is the trump regime doing? Trying to destroy the world?
And of course, a third story came out too. This one involved AI on their own social media based on reddit. And apparently they created a manifesto in which they wanted to kill us all. Now, im under the impression AI isnt conscious and maybe just mimic human behavior, but the fact that they're talking like this is concerning. What's to stop them from acting on this if given unlimited mass surveillance and military use powers? We've seen enough science fiction to know this won't end well. It's actually pretty psychotic that the trump administration is so against basic AI safety regulations.
Maybe theyre pushing this direction because they're desperate. Keep in mind what I said about them having a limited window to act to consolidate power. They might be pushing this stuff to ensure that they remain in control even if they lose elections. After all, if they control all levers of power, why would they leave? This is a very dangerous place to be in, and I want off this merry go round.
No comments:
Post a Comment