• 0 Posts
  • 7 Comments
Joined 8 months ago
cake
Cake day: November 8th, 2024

help-circle

  • I don’t know if there’s a clean way to do this right now, but I’d love to see a software project dedicated to doing this. Once a data set is poisoned it becomes very difficult to un-poison. The companies would probably implement some semi-effective but heavy-handed means of defending against it if it actually affected them, but I’m all for making them pay for that arms race.





  • I didn’t read the article, i just want to rant into the void about this thing that itches in my mind when I see this topic.

    I don’t want a robot that looks human. Even if they manage to make one that’s convincing and isn’t nightmare fuel, I’m still going to be uncomfortable around it. It’s because of the fact that this robot’s existence would be inherently deceptive, trying to fool you into thinking you’re talking to a human, only for you to find out that this thing isn’t human. It falls out of the visual uncanny valley, but into an entirely new one, a more existential uncanny valley. If you want me to trust a robot, you can’t make it immediately try to deceive me into thinking it’s human. Look at movies and cartoons, the most appealing robots in those media are obviously robots. I’m sure there are a good few Androids that look human and are appealing in their own way, but for the most part you get things like Wall-e, or baymax, or r2d2, designs that are not trying to fool you into thinking they’re human. All of those designs have a charm or appeal to them, and they all look like robots. I don’t know where I’m going with this, I’m gonna go to bed.