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...
Hi Mahendra — we're hoping to get a post out about this shortly, sorry for the delay here. The next EAG Bay Area will take place from Feb 2–4, 2024. We're planning for this event to have a more x-risk/global catastrophic risk focus than our standard EAGs, though will explain this in more detail in our upcoming post.
Thanks Eli.