Congrats to the team on the metrics stabilizing and also congrats Sarah on the promotion!^[1]

1. ^{^}

   Also maybe congrats to Toby, although I can't tell if the new title is a promotion?

Those seem like reasonable categories. I disagree with the antecedent but will rephrase  your question as "what type of decline am I most worried about?", the answer to which is "1".

In particular: I think it's possible that AI will be such a compelling issue that the intellectual excitement and labor moves away from cause-neutral EA to EA-flavored AI Safety. This is already happening a bit (and also happened with animal welfare); it doesn't seem crazy to think that it will become a larger problem.

Thanks for cross posting these blogs to the forum. You are one of my favorite new (to me) EA writers, and I probably wouldn't see your work as much if it wasn't cross posted.

Thanks for writing this, I think it's interesting to read case studies like these. 

Recovery is not likely historically

As with many "p->q" arguments, I think this one is true for the trivial reason that q holds independent of p. I.e. it's true that it's unlikely that people will identify as EA's one hundred years from now^[1] but that also was unlikely 5 years ago. 

I asked Claude "What are five randomly chosen social movements which started at least 100 years ago? Give no regard to whether they still currently exist." and it said:

The Temperance Movement (1820s-1930s) - A campaign to reduce or eliminate alcohol consumption, which gained significant momentum in the United States and other countries. It ultimately led to Prohibition in the US from 1920-1933.

The Suffragette Movement (1840s-1920s) - The organized effort to secure voting rights for women, with major campaigns in Britain, the United States, and other countries. It employed tactics ranging from peaceful protests to more militant actions.

The Abolitionist Movement (1830s-1870s) - The campaign to end slavery, particularly prominent in the United States and Britain. It included figures like Frederick Douglass and William Lloyd Garrison and contributed to the American Civil War.

The Eight-Hour Day Movement (1860s-1880s) - Labor activism focused on establishing an eight-hour working day, with major events like the Haymarket Affair in Chicago in 1886. It was part of broader labor organizing efforts worldwide.

The Theosophical Movement (1875-early 1900s) - A spiritual and philosophical movement founded by Helena Blavatsky and others, blending Eastern and Western religious concepts. It promoted ideas about universal brotherhood and ancient wisdom traditions.

None of these still exist, though three (suffragette, abolitionist, eight-hour day) ~accomplished their goals which maybe you want to exclude as a reason for movements to stop existing. So I asked for three more and it said:

The Back-to-Africa Movement (1890s-1920s) - Led by figures like Marcus Garvey and his Universal Negro Improvement Association, this movement encouraged people of African descent to return to Africa and promoted Black nationalism and pride. It was particularly influential in the early 20th century.

The Settlement House Movement (1880s-1920s) - A social reform effort where educated volunteers lived in poor urban neighborhoods to provide services and advocate for better conditions. Jane Addams' Hull House in Chicago was one of the most famous examples.

The Eugenics Movement (1900s-1940s) - A pseudoscientific social movement that promoted "improving" human heredity through selective breeding and sterilization programs. While thoroughly discredited today, it was influential in early 20th century policy in many countries including the United States and Germany.

So the base rate for lasting 100+ years is either 0/8, 0/5, or 3/8, depending on how you want to classify the movements which accomplished their goals. This does not seem far off from the 1/5 success rate in your data set.

1. ^{^}

    I roughly mentally tried to average the age of your case studies and thought that they were about 100 years old on average, but this isn't precise.

Fixed the link. I also tried your original prompt and it worked for me.

But interesting! The "Harder word, much vaguer clue" seems to prompt it to not actually play hangman and instead antagonistically try to post hoc create a word after each guess which makes your guess wrong. I asked "Did you come up with a word when you first told me the number of letters or are you changing it after each guess?" And it said "I picked the word up front when I told you it was 10 letters long, and I haven’t changed it since. You’re playing against that same secret word the whole time." (Despite me being able to see its reasoning trace that this is not what it's doing.) When I say I give up it says "I’m sorry—I actually lost track of the word I’d originally picked and can’t accurately reveal it now." (Because it realized that there was no word consistent with its clues, as you noted.)

So I don't think it's correct to say that it doesn't know how to play hangman. (It knows, as you noted yourself.) It just wants so badly to make you lose that it lies about the word.

Huh interesting, I just tried that direction and it worked fine as well. This isn't super important but if you wanted to share the conversation I'd be interested to see the prompt you used.

By analogy, o4-mini's inability to play hangman is a sign that it's far from artificial general intelligence (AGI)

What is your source for this? I just tried and it played hangman just fine.

Given that some positions in EA leadership are already elected, I might suggest changing the wording to something like:

There should be an international body whose power is roughly comparable to CEA whose leadership is elected

I think I agree with your overall point but some counterexamples:

1. EA Criticism and Red Teaming Contest winners. E.g. GiveWell said "We believe HLI’s feedback is likely to change some of our funding recommendations, at least marginally, and perhaps more importantly improve our decision-making across multiple interventions"
2. GiveWell said of their Change Our Mind contest "To give a general sense of the magnitude of the changes we currently anticipate, our best guess is that Matthew Romer and Paul Romer Present's entry will change our estimate of the cost-effectiveness of Dispensers for Safe Water by very roughly 5 to 10% and that Noah Haber's entry may lead to an overall shift in how we account for uncertainty (but it's too early to say how it would impact any given intervention)."
3. HLI discussed some meaningful ways they changed as the result of criticism here.