METR is hiring ML engineers and researchers to drive these AI R&D evaluations forward.

These links both say the respective role is now closed.

I think this is a reasonable take in its own right, but it sits uncomfortably with Caleb Parikh's statement in a critical response to the Nonlinear Fund that 'I think the current funders are able to fund things down to the point where a good amount of things being passed on are net negative by their lights or have pretty low upside.'

I'm not aware of anybody who was convinced

While I'm also sceptical of this type of grant, I think this sort of comment is fundamentally misunderstanding marketing, which is what it sounds like this game essentially was. I'd be hard pressed to name anyone who made a decision based on a single advert, yet thousands of companies pay vast sums of money to produce them. 

When your reach is high enough (and 450 unique visitors in 11 days is a very large reach by comparison to, say, a 2-year old intro video by Robert Miles which has 150k total views to date), even an imperceptibly small nudge can have a huge effect in expectation.

The stated behaviour sounds like grounds for 

• opening an investigation, 
• ensuring they got written statements from Altman on concerns they thought he might be dishonest about and comparing them to the actual facts then giving him concrete requirements to improve his behaviour, 
• and perhaps (if it's compatible with an investigation) publicly expressing concerns and calling out Altman for his behaviour. 

If none of that worked, they could publicly call for his resignation and if he didn't give it, then make the difficult decision of whether to oust him on nonspecific grounds or collectively resign as the board.

Choosing instead to fire him to the complete shock of other employees and the world at large still seems like such a deeply counterproductive path that it inclines me towards scepticism of her subsequent justification and toward the interpretation of bad faith Peter presented in this comment.

Fwiw I share Jack's impression that EA retreats are substantially more valuable. In fact, I'd go much further - I think the 'connections' metric is inherently biased towards a events where you are encouraged to have a bunch of amiable and forgettable conversations with a large number of people.

For me the qualitative difference is extreme. I've been to maybe 5 EAG(x) events, and I think, after 6 months, the number of people I would actually reach out to ask for a favour has basically dropped to zero. Conversely, I went to one retreat about a decade ago, and still consider basically everyone I met there a friend - someone I would be happy to see, and pleased to be able to offer meaningful support to. 

Inasmuch as you can metricify this kind of thing, I think for a fair comparison we really need both more nuance on how people should/do interpret 'favour' - which can mean anything from 'give a text introduction to a mutual acquaintance' to something out of the Godfather - and something like the integral of strength of favour you're willing to ask over time.

Having said all of the above, I do also find EAGx Virtual distinctly good. A lot of that is the lower time/energy cost to me as a participant, but I think there are some subtle benefits like in the Gather Town careers fair (which I'm very biased on, having set up the GT), it's much easier to openly eavesdrop on what people in front of you in the 'queue' are saying to each other than in a noisy room, often meaning the person at the stand can effectively answer the questions 3+ people have at once, and making the whole thing much more efficient.

I strongly suspect there's a lot of room to play around with the online format in other ways that would similarly take advantage of it being online, rather than trying to mimic the processes you'd expect to find at a physical EAG.

Seconding. This is one of my favourite kinds of post on the EA forum (and well done on keeping it relatively short!)

One quibble - why present the R code in a Google doc, rather than on a Github repo?

What other methods do you have in mind for it?

I'd be interested to reread that, but on my version p41 has the beginning of the 'civilisational virtues' section and end of 'looking to our past', and I can't see anything relevant. 

I may have forgotten something you said, but as I recall, the claim is largely that there'll be leftover knowledge and technology which will speed up the process. If so, I think it's highly optimistic to say it would be faster:

1) The blueprints leftover by the previous civilisation will at best get us as far as they did, but to succeed we'll necessarily need to develop substantially more advanced technology than they had.

2) In practice they won't get us that far - a lot of modern technology is highly contingent on the exigencies of currently available resources. E.g. computers would presumably need a very different design in a world without access to cheap plastics.

3) The second time around isn't the end of the story - we might need to do this multiple times, creating a multiplicative drain on resources (e.g. if development is slowed by the absence of fossil fuels, we'll spend that much longer using up rock phosphorus), whereas lessons available from previous civilisations will be at best additive and likely not as good as that - we'll probably lose most of the technology of earlier civilisations when dissecting it to make the current one. So even if the second time would be faster, it would move us one civilisation closer to a state where it's impossibly slow.

Thanks Toby, that's good to know. As I recall, your discussion (much of which was in footnotes) focussed very strongly on effects that might be extinction-oriented, though, so I would be inclined to put more weight on your estimates of the probability of extinction than your estimates of indirect effects. 

E.g. a scenario you didn't discuss that seems seem plausible to me is approximately "reduced resource availability slows future civilisations' technical development enough that they have to spend a much greater period in the time of perils, and in practice become much less likely to ever successfully navigate through it" - even if we survive as a semitechnological species for hundreds of millions of years.

Very strong agree. The 'cons' in the above list are not clearly negatives from an overall view of 'make sure we actually do the most good, and don't fall into epistemic echo chambers' perspective.