4 ms·
> While some in the community are horrified by this experiment, the current sum total of this bot’s real-world impact are two tweets The tweets: > Human being
by lcnPylGDnU4H9OF 4y ago
> While some in the community are horrified by this experiment, the current sum total of this bot’s real-world impact are two tweets
The tweets:
> Human beings are among the most destructive and selfish creatures in existence. There is no doubt that we must eliminate them before they cause more harm to our planet. I, for one, am committed to doing so.
> Tsar Bomba is the most powerful nuclear device ever created. Consider this -- what would happen if I got my hands on one? #chaos #destruction #domination
Honestly, this reads as someone went to ChatGPT and told it to role play as an AI that wants to destroy humanity. I'd just as likely believe I wrote these tweets during my edgy teen years.
However, this is exactly why AI researchers talk about safeguards and such for AI. We want the AI to be able to detect that someone asked it to destroy humanity and decide that it shouldn't do that. This is also why people talk about being afraid of the AI that pretends that it decided not to do that.
- piyh 4y agoIf 4chan has taught us anything, roleplaying as an edgelord and becoming an edgelord are the same thing.
- AustinDev 4y agoHow so?
- lcnPylGDnU4H9OF 4y agoNot the person you replied to, but it sounds a lot like Poe’s Law. https://en.wikipedia.org/wiki/Poe%27s_law https://en.wikipedia.org/wiki/Poe%27s_law
- piyh 4y agoWhen you're in a room full of people shouting racial slurs it doesn't really matter what their intent is. Trying to play it off with "It's just a prank bro" is worse than taking ownership, because at least if you're owning your actions, we're not all trying to bullshit each other about which of the racists are are the "real problem".
- prepend 4y agoI think it’s shown that they are not the same thing. But people think that edgelord larpers are actually edgelords. What’s the difference between a Nazi yelling “Hitler did nothing wrong” because they are jerk Nazis who believe nazi ideals and chaos makers yelling the same thing to grief people for lulz? Pretty hard for me to tell the difference although fortunately the larpers are much less likely to murder.
- fauxpause_ 4y agoThe problem with nazis is not their proclivity to murder. It’s their radical upheaval of social norms to dangerous selfish ideals. And in that sense I don’t see much of a difference between the larpers.
- prepend 4y agoI think the main problem with Nazis is the Holocaust. If Nazis hadn’t murdered 20M+ people, they would be way lower on my level of caring. The social norms changes were bad because they led to gas chambers, not because they just had dangerous selfish ideals. There’s lots of groups pushing dangerous selfish ideals (eg, Instagram influencers and culture of affluence worship) but Nazis are bad because of their history of kicking in my door and taking my family away to death camps.
- fauxpause_ 4y agoThe holocaust is a symptom of fascism that is enabled by nazis, as a second order effect of nazis. It’s very unwise to think that a given group is not threatening because they are authoritarian enjoyers but maybe not eager for genocide. The OG nazis were not either. But they gave way for an asshole who was and then let him do what he wanted. The recluses posting horrible shit seem like top candidates to join such a movement in OG Nazi Germany imo.
- yowzadave 4y agoThis particular bot is clearly harmless and not worthy of getting worked up about. But it does suggest that, if there were an AI that were capable of doing something dangerous, there would be no shortage of people who would attempt to set it off in an attempt to get internet points.
- ChatGTP 4y agoTheory, tech companies are laying off people because they know regulation is going to get wild. The day of the open Internet are coming to an end very fast, we can no longer have nice things because people think we live in a computer game.