Some GMU economists (e.g. @robinhanson @tylercowen) are much less worried than @ESYudkowsky or @sama about x-risk from powerful AI.

My attempt to steelman one of their main arguments:

1. The economy is a complex system with thousands of hard bottlenecks & fragile dependencies. 2. This greatly limits the power of AI systems—a bit like how CEOs and US Presidents are much less powerful than they seem.

3. This means we'll have a lot of time to notice AI systems that are trying to mess with us.

This month, Joe Carlsmith published what is, for me, his best essay yet.

I now realise, more clearly than I have before, just how much my metaphilosophy is shaped by the novels of Virginia Woolf, which I read and re-read in my youth.

joecarlsmith.com/2023/02/16/why… As Nietzsche pointed out, moral realists often have a clear Platonic sensibility, which seeks eternal truth rather than contingent or marginal insights.

If you seek that kind of thing, your arguments need the force of a proof: logical inference from solid axioms.

Some people think @sama is ignorant or reckless.

In February 2015, 10 months before founding @OpenAI, he wrote a blog post which began:

“Development of superhuman machine intelligence is probably the greatest threat to the continued existence of humanity.” The post contains a lot of vintage Bostrom, and strongly recommends Superintelligence.

So: he’s not ignorant!

Reckless? That one is much harder to settle. But… if you think his plan is obviously stupid, I’m pretty sure you’re underestimating him.

blog.samaltman.com/machine-intell…

Friedrich Nietzsche on effective altruism, according to ChatGPT.

1. "Effective altruism is the last refuge of the ascetic ideal, a way for the weak-willed to feel virtuous without actually having to face the difficult and messy realities of life. It is a way for the timid to assuage their guilt and avoid the uncomfortable task of self-examination. True compassion and strength come from embracing life in all its complexity, not from fleeing from it through shallow acts of charity."

My vote for “effective altruism CEO of the year” goes to @peterwildeford.

1. Over the past couple years @rethinkpriorities have grown their research teams faster than any other EA org. Their research database is, increasingly, an absolute banger. 2. Their rate of improvement has been faster than any other EA research org.

When they started, some influential figures doubted their early team, indulging in some “devalue & dismiss”.

These figures did not recognise the nascent talent of @peterwilderford. He pushed on anyway.

Peter Thiel explains how @ESYudkowsky & @MIRIBerkeley went from transhumanism to luddite despair, in the course of 15 years.

From alignment research to death with dignity...

1/15 Thiel: "Nick Bostrom, he's an Oxford academic... most of these people are interesting because they have nothing to say. They're just mouthpieces, it's like the mouth of Sauron... they're useful because they tell us where the zeitgeist is."

2/15

Derek Parfit had a great passion for photography, as well as philosophy.

As reported by the Financial Times:

“For more than 20 years, Derek Parfit would spend five weeks every winter in Venice and St Petersburg taking photographs. He returned obsessively to the same places…” “… in Venice, Palladio’s churches, the Doge’s Palace and the Grand Canal; in St Petersburg, the Winter Palace and the General Staff Building.

Although Parfit shot thousands of rolls of film over more than two decades…”

Who and what rises in status due to the collapse of FTX?

Ten things that rise in status, in my eyes:

1. The pragmatism of Fat Tony, as described by @nntaleb. Fat Tony mostly ignores abstract arguments. He rarely even thinks in such terms. He prefers the time-tested rules of common sense.

Moral universalism is a one-way ticket to crazy town: take the ride—but not too far.

Some effective altruist memes may not scale gracefully.

What does the collapse of FTX mean for effective altruism?

@SamoBurja has the best analysis I've seen so far.

Firstly, writes Samo: "The moral authority and intellectual legitimacy of EA will be reduced."

1/12 Before I continue: a content warning, and some context.

2/12

"Derisive and contemptuous in tone, yet weak in argument". Jeff McMahan (2016) on the remarkably poor quality of philosophical critiques of effective altruism. forum.effectivealtruism.org/posts/hvYvH6wa… Effective altruism starts with widely held moral intuitions. It is not dependent on utilitarianism. (See e.g. forum.effectivealtruism.org/posts/av7MiEhi…)

In 2001, @robinhanson shared 14 wild ideas, of which he thinks at least a third are true. My favourites...

1. Many times each day, your mind permanently splits into different versions that live in different worlds. The "many worlds" interpretation of quantum mechanics says every possible quantum outcome actually happens in a different "world." [...] QM is our most basic theory of physics, and surveys of prominent physicists reportedly find majorities favoring this interpretation.

Some assumptions of effective altruism, according to me:
forum.effectivealtruism.org/posts/av7MiEhi… 1. Two-thirds utilitarianism. Utilitarianism is a useful and underrated way to think about what matters in some circumstances. Other theories of value and normative frameworks should be given serious consideration and weight, partly due to moral uncertainty. Taking utilitarianism