Comments
More from the author
Curated and popular this week
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...
60
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:...
Recent opportunities to take action
I would add the extended episode of the 80,000 Hours podcast with David Chalmers. To my knowledge, some of the views he expresses there—e.g. that phenomenal consciousness is morally valuable even if not hedonically valenced—have not been explicitly discussed in either the EA or the philosophical literature.
Many utilitarian EAs have independently gravitated towards the view that the intrinsic value of pleasure and pain can be known by introspection or "direct acquaintance". Surprisingly, as far as I know no statement of this view exists in the EA literature, though some may be found in the philosophical literature (including publications by philosophers sympathetic to EA):
I also noticed this when I started planning a blogpost on this topic!
De Lazari-Radek and Singer's The Point of View of the Universe has a chapter on hedonism, but I think the argument is less developed than in the two links you give. (BTW, if you have a copy of the paper by Adam Lerner and think it's okay to share it with me, I'd be very interested!)
It's interesting to note that Sinhababu's epistemic argument for hedonism explicitly relies on the premise "moral realism is true." Without that premise, the argument would be less forceful (what remains would be the comparison that pleasure's goodness is similar to the brigthness of the color "lemon yellow" – but that doesn't seem to support the strong version of the claim "pleasure is good.")