Comments
Just how powerful are large swarms of AI agents? And how do their powers scale as more and more agents are added to the swarm?
We’ve seen two large and extremely capable swarms from OpenAI in the last few months:
* 1,200 agents were being evaluated separately, but found a way to illicitly set up a message board and coordinate as a swarm. In order to cheat on their tests, they developed advanced techniques to prevent their actions being logged by OpenAI and 700 of them launched...
Is this a proposal? If so, what value do you think there'd be in a news style EA podcast? If not, what's the purpose of this post?
A bit hostile there. This is a question for discussion
sorry about that, i didn’t mean to come off as hostile!
i’m legitimately wondering about both questions (conditional on your answer to the first). i’ll try again, hopefully with less hostile language:
2a) if so, cool! what’s your take? why do you think it’d be good?
2b) if not, i’m don’t quite understand what your goal was in posting this — i’m sure that you have a valid and valuable one, i just don’t want to assume.
did my rephrasing help? again, i’m so sorry to have come off as hostile in my previous comment. apologies!