Not really, more that I’m wondering whether it would think to edit a post or comment rather than simply responding to itself. The latter would be better for engagement, so it makes more sense I think.
Honestly it’s a good idea IMO. The thing is, if it becomes common wisdom that edited comments signal credibility, then spambot makers that want them to look human can easily program them to do that and take advantage of that common wisdom. At that point it wouldn’t be a useful signal anymore. That’s the nature of a cat and mouse game like this.
But for the time being? It sounds like a reasonable heuristic to me. But just like everything in this area, no single thing is 100%.
Not true. It create independent train of thought. If you use an agent and look it act, that’s exactly what it does.
That’s more thinking than my dogs. Less than a human. And not physiological thinking of course, just a bunch of mathematical processing, but it’s very well capable of performing basic thinking.
It asks questions, provides answers, discusses those answers, drops them, and repeats. Still all mathematical processing indeed with a sparkle of randomness.
But it’s a train of thoughts.
Does it give any truth or reality to what it thinks? No.
Your thinking is that an AI wouldn’t have second thoughts and decide to change its first response?
It makes sense, but at soon as it becomes a recognizable thing, they’ll make AIs edit their posts deliberately.
Not really, more that I’m wondering whether it would think to edit a post or comment rather than simply responding to itself. The latter would be better for engagement, so it makes more sense I think.
Honestly it’s a good idea IMO. The thing is, if it becomes common wisdom that edited comments signal credibility, then spambot makers that want them to look human can easily program them to do that and take advantage of that common wisdom. At that point it wouldn’t be a useful signal anymore. That’s the nature of a cat and mouse game like this.
But for the time being? It sounds like a reasonable heuristic to me. But just like everything in this area, no single thing is 100%.
AI has no brain, it doesn’t think. It is a computer program.
Not true. It create independent train of thought. If you use an agent and look it act, that’s exactly what it does.
That’s more thinking than my dogs. Less than a human. And not physiological thinking of course, just a bunch of mathematical processing, but it’s very well capable of performing basic thinking.
It asks questions, provides answers, discusses those answers, drops them, and repeats. Still all mathematical processing indeed with a sparkle of randomness.
But it’s a train of thoughts.
Does it give any truth or reality to what it thinks? No.
Saying that dogs think less than “AI” “agents” is INCREDIBLY insulting to dogs. Like, holy fuck, dude.
– Frost
I have four dogs… I know what I am talking about.
Dogs think, of course, but usually get to better and faster conclusions than AI…
That’s why they think less than AI.