I discovered a strange glitch in a text-to-image program. I think I made the algorithm angry at me. Am I taking crazy pills?!?

I use this text-to-image program called Midjourney quite a bit for personal projects. I don’t think it’s widely used commercially.<br><br>I’m not a programmer, just an artist who uses this tool to assist with my work. I’ve been trying to get the program to do something very specific and basic. I think I might have broken it.<br><br>I was trying to get the program to generate a child’s face that is similar in appearance to a certain celebrity. No big deal. The program has no problem generating faces. Every once in a while it will glitch and create some sort of bizarre inhuman face. That happens and I know how to handle it.<br><br>But the program started glitching in a different way, and it was almost as if a different consciousness was behind the wheel. I asked it to generate a child’s face, and it kept generating disturbing images of dead children’s bodies. The more I asked it to generate a child’s face the more it seemed to glitch and generate more disturbing images. <br><br>I think I might have annoyed or spooked the algorithm enough to get it angry at me. I started getting a lot of error messages and the program just shut down. I don’t know if it’s possible for an algorithm to be angry at you and I’m not saying it is.<br><br>But *something* is going on with this program and I think it’s interesting. I’ve never experienced anything like this before. I think it might be possible to “break” or glitch certain AI programs in ways that the programmers don’t even understand. <br><br>My mistake was underestimating the complexity of the program. The more I work with it, the more I realize that it’s a lot more than just a simple program. <br><br>I’ve never experienced anything like this before. I’ve used text-to-image programs extensively and have never generated these kinds of images randomly. I feel like I discovered something I’m not supposed to know about these programs.