Aidan Alexander
Posts 10
Comments82
I have a somewhat basic question: If I'm currently of the view that OpenAI is reckless and I'd choose for them to stop existing if I could, does it follow that I should want Musk to win this? Or at least that I should want OpenAI to fail to restructure?
(Also, if you want OpenAI to successfully restructure, I'd love to know why. Is it just a "we need to US to win the race with China" thing?)
From first hand experience, I think critics should make more of an effort to ensure that the charities have actually received your communications and had a chance to review it. When my organization was the subject of a critical post, the email from the critic had landed in my spam, so I only learned of the critique when reading it on the forum.
What's more, as an organization of 2 people, one of which was on leave at the time of the notice, we couldn't have reasonably responded to it without stopping critical work to keep the lights on. The smaller the organization being critiqued is, the less flex capacity they have to respond quickly to these sorts of things, and so they should be given a longer grace period.
I can't believe how often I have to explain this to people on the forum: Speaking with scientific precision makes for writing very few people are willing to read. Using colloquial, simple language is often appropriate, even if it's not maximally precise. In fact, maximally precise doesn't even exist -- we always have to decide how detailed and complete a picture to paint.
If you're speaking to a PhD physicist, then say "electron transport occurs via quantum tunneling of delocalized wavefunctions through the crystalline lattice's conduction band, with drift velocities typically orders of magnitude below c", but if you're speaking to high-school students teetering on the edge of losing interest, it makes more sense to say "electrons flow through the wire at the speed of light!". This isn't deception -- it's good communication.
You can quibble that maybe charities should say "may" or "could" instead of "will". Fine. But to characterize it as a wilful deception is mistaken.
If charities only spoke the way some people on the forum wish they would, they would get a fraction of the attention, a fraction of the donations, and be able to have a fraction of the impact. You'll get fired as a copywriter very quickly if you try to have your snappy call to action say "we have estimates that state every $1 you donate will spare 1,770 piglets from painful mutilations".
I think it's pretty safe to assume that the reality of most charities' cost-effectiveness is less than they claim.
I'd also advise skepticism of a critic who doesn't attempt to engage with the charity to make sure they're fully informed before releasing a scathing review. [I also saw signs of naive "cost-effectiveness analysis goes brrr" style thinking about charity evaluation from their ACE review, which makes me more doubtful of their work].
It's also worth noting that quantifying charity impact is messy work, especially in the animal cause area. We should expect people to come to quite different conclusions and be comfortable with that. FarmKind estimated the cost-effectiveness of Sinergia's pig work using the same data as ACE and came to a number of animals helped per dollar that was ~6x lower (but still a crazy number of pigs per dollar). Granted, the difference between ACE and Vetted Causes assessments are beyond the acceptable margin of error
I am also skeptical of small percentages but more so because I think that the kinds of probability estimates that are close to 0 or 1 tend to be a lot more uncertain (perhaps because they’re based on rare or unprecedented events that have only been observed a few times).
I’m no statistician, but I’m not sure that we can say that small percentages tend to be exaggerated though… For one, I recall reading in Superforecasters that there’s evidence that people tend to underestimate the likelihood of rare events and overestimate the likelihood of common ones in forecasting exercises, so that’s an piece of evidence pointing towards small probabilities generally being too low rather than too high. Secondly, a low probability can equally be framed as a high probability of that event not happening. So in short - I agree that probability estimates that are close to 0 or 1 tend to be less certain, but not that probability estimates close to 0 tend to be overestimates any more than underestimates
Far from convinced that continued existence at currently likely wellbeing levels is a good thing