Comments by "Anony Mousse" (@anon_y_mousse) on "Why Training AI like GPT-4/ChatGPT to Act Human is Ultra Dangerous" video.

  1. It's precisely because of the people working on it that it will be very dangerous. People have already used it to commit fraud, it's only a matter of time before it's misused to kill a great many people, especially those that can tell the difference between a man and a woman. Consider how far computer vision programs have come. They've gotten so good that they can recognize a cat rolling around on the floor as a cat, and it's not long before they start using it in drones. Feed it pictures of protestors and no more political prisoners protesting fraudulent elections. Those on the left are very violent and evil. Given the number of people either bending over backwards to excuse murdering children or outright celebrating it, especially because they were attending a Christian school, I don't see a good future with this technology in play.
    1
  2. I've come to much the same conclusion. It's not true AI, what people often refer to as AGI, but it is a sophisticated tool that can and will be misused in ways that will likely lead to a lot of people dying. It's already been used to commit fraud after all.
    1