guajalote
5d ago
It's clear to me that aligned AI (of the type you describe) is a much bigger threat in the near term than unaligned AI. I haven't seen any indication that AI intrinsically wants to harm people, but an aligned AI will gladly do bad things because a human told it to.
guajalote
5d ago
I'm not claiming it's what she meant. She probably didn't mean anything or even write the speech. I'm just pointing out how a small tweak by the speechwriter would have saved the day.
guajalote
6d ago
One of the things that makes it so hard to align superintelligent human beings (like political leaders) is that they have human drives and ambitions for things like sex, power, prestige, etc. They will always be intrinsically driven to get around guardrails to get these things. It doesn't seem like AI needs to be beholden to its goals in the same way. It seems like we can design a paperclip maximizer that just stops maximizing paperclips when we tell it to. Hopefully that continues to be true.
guajalote
6d ago
I agree with you philosophically. I have a son the same age who is healthy, but if he were suffering severely in the manner described in the article I would feel compelled to at least seriously consider every lawful option to alleviate that suffering.
However, I think there might be good practical arguments against euthanasia, especially in a system where the government bears the cost of healthcare. There is a strong financial incentive for the government to use euthanasia as cost-saving tool, and this incentive might erode whatever checks are in place, such that people who should not be euthanized are getting euthanized anyway. I'm not sure if I fully believe this argument, but it's the only counterargument that seems at all convincing to me.
guajalote
6d ago
Or if she had said "the primary victims of the aftermath of war" which seems to be what she is trying to say in the following sentences.
guajalote
6d ago
Its extremely rarely that I agree with China or Trump on anything, but I think they are sadly correct. No matter how well-meaning and theoretically prudent AI safety regulation may be, the practical effect of any real world AI regulation will not be safety, but will be to "uphold Washington's monopolistic hegemony in cutting-edge technology, and exclude China from the global AI governance system."
guajalote
26d ago
I don't see how that pertains to the law being discussed. Of course private venues can turn people down. The law bans drag shows that were not turned down by the venue.
guajalote
26d ago
It makes for an incredibly cheesy, campy effect, and I'm tired of pretending that it's not in horrible taste.
Speech that's cheesy, campy, and in horrible taste is protected by the First Amendment.
guajalote
26d ago
This argument is like saying that if an impersonation of Barack Obama by a white man is racist, then being Barack Obama is racist.
Yes, this argument would be equally bad. An impersonation of Barack Obama by a white man is not inherently racist. Also, racist speech is protected by the First Amendment.
In many contexts the act of impersonation itself is sexual, above and beyond the fact that there's sexuality because of the target of the impersonation.
I'm not sure this is true. But even if it were true, there are many contexts where impersonation is not intrinsically sexual, and the law provides no mechanism to distinguish these contexts. The law does not even mention impersonation as a factor to be considered in the analysis.
One of the common responses to AI safety concerns is "just don't let it get out of the box." That's actually a pretty good point: as long as the AI doesn't have godlike intelligence and as long as we actually take steps to keep it inside the box where it can't impact the real world directly, the danger that an AI can pose is somewhat contained and limited.
When you start giving AI access to real world stuff, you no longer have the ability to keep it fully inside the box. At this point it's probably fine and not a big risk given current AI capabilities, but it's hard to say for sure, and it gets riskier and riskier as capabilities get better.
More options
Context Copy link