Not malevolent, indifferent. There are vastly more possible goals that lead to everyone dying than there are of something that takes account of human values at all whether to maximise negative or positive utility in some fashion. AI safety was one of the first EA causes but it’s hardly hijacked the community, just look at the last ten episodes of the podcast. One of the ways to be effective is to focus on neglected problems, and avoiding “everyone dies” even with small probability may have a lot of money allocated to it but it’s not enough and it’s not distributed by expected value, at all.
> #57 – Tom Kalil on how to do the most good in government
> #56 – Persis Eskander on wild animal welfare and what, if anything, to do about it
> #55 – Mark Lutter & Tamara Winter on founding charter cities with outstanding governance to end poverty
> #54 – Askell, Brundage & Clark from OpenAI on publication norms, malicious uses of AI, and general-purpose learning algorithms
> #53 – Kelsey Piper on the room for important advocacy within journalism
> #52 – Prof Glen Weyl on uprooting capitalism and democracy for a just society
> #51 – Martin Gurri on the revolt of the public & crisis of authority in the information age
> #50 – Dr David Denkenberger on how to feed all 8 billion people through an asteroid/nuclear winter
> #49 – Dr Rachel Glennerster on a year's worth of education for under $1 and other development best buys
> #48 – Brian Christian on better living through the wisdom of computer science
> #47 – Catherine Olsson & Daniel Ziegler on the fast path into high-impact ML engineering roles
The point is there's zero evidence to show "AI kills everyone" is even a small risk.
By your logic, my plan to dedicate $100m to making offerings to Beelzebub to spare humanity is outstanding value for money dedicated to a neglected long-tail risk. Beelzebub has very few worshipers, and the odds of his choosing to exterminate mankind are low, but if he did, it would be very bad for us.
> #57 – Tom Kalil on how to do the most good in government
> #56 – Persis Eskander on wild animal welfare and what, if anything, to do about it
> #55 – Mark Lutter & Tamara Winter on founding charter cities with outstanding governance to end poverty
> #54 – Askell, Brundage & Clark from OpenAI on publication norms, malicious uses of AI, and general-purpose learning algorithms
> #53 – Kelsey Piper on the room for important advocacy within journalism
> #52 – Prof Glen Weyl on uprooting capitalism and democracy for a just society
> #51 – Martin Gurri on the revolt of the public & crisis of authority in the information age
> #50 – Dr David Denkenberger on how to feed all 8 billion people through an asteroid/nuclear winter
> #49 – Dr Rachel Glennerster on a year's worth of education for under $1 and other development best buys
> #48 – Brian Christian on better living through the wisdom of computer science
> #47 – Catherine Olsson & Daniel Ziegler on the fast path into high-impact ML engineering roles