One thing this makes me curious about: how good is the existing evidence base on electric stunning is better for the welfare of the shrimp, and how much better is stunning? I didn't realize SWP was thinking of using the corporate campaign playbook to scale up stunning, so it makes me curious how robustly good this intervention is, and I couldn't quickly figure this out from the Forum / website. @Aaron Boddy🔸 is there a public thing I can read by any chance? No pressure!
FWIW, "how good is stunning for welfare" is the main question I have about the impact of HSI right now (whereas other crucial considerations like "how many shrimp do you target", "will farmers use the stunners", "how should we value shrimp / their expected sentience" all feel clearer to me).
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...
TLDR: Everyone’s talking about what the money could do, but few about how to decide where it goes.
This post is part of the new series of articles on cross-cause giving and the new wave of philanthropy. Stay tuned to the EA Forum and our Substack for the latest takes on topics such as giving now vs. later, common pitfalls in cause prioritization, and other crucial considerations from the Cross-Cause Fund (CCF) team...
Summary:
First, I give several different angles on how I feel about reinforcement learning:
* Theoretical case: RL is a black-box source of agency — this should give us classic misalignment worries, especially compared to agency-via-scaffolding
* Recent incidents (huggingface etc) and more mundane forms of misaligned behaviour in personal use give me bad vibes about the direction-of-travel of recent AI progress
* I’m worried things might get worse:...
Well done to the Shrimp Welfare Project for contributing to Waitrose's pledge to stun 100% of their warm water shrimps by the end of 2026, and for getting media coverage in a prominent newspaper (this article is currently on the front page of the website): Waitrose to stop selling suffocated farmed prawns, as campaigners say they feel pain
This is so great, congratulations team!! <3
One thing this makes me curious about: how good is the existing evidence base on electric stunning is better for the welfare of the shrimp, and how much better is stunning? I didn't realize SWP was thinking of using the corporate campaign playbook to scale up stunning, so it makes me curious how robustly good this intervention is, and I couldn't quickly figure this out from the Forum / website. @Aaron Boddy🔸 is there a public thing I can read by any chance? No pressure!
FWIW, "how good is stunning for welfare" is the main question I have about the impact of HSI right now (whereas other crucial considerations like "how many shrimp do you target", "will farmers use the stunners", "how should we value shrimp / their expected sentience" all feel clearer to me).
Not sure if it’s exactly what you’re looking for, but this post seems relevant.