AI safety and the Last Man
Marc Andreessen, Wolf Tivvy and Nick Land all think that the AI safety crowd exhibit the fearfulness of Nietzsche’s Last Man in the face of “dangerous agency” (Wolf’s phrase) 1.
The Last Man dreams of a procedural utilitarian peace and security. He hopes that moral philosophy and rational decision theory might lead to an end of war.
By contrast: the Nietzschean, evolutionary thinkers valorise struggle, bravery and Great Men. For them, Darwinian competition continues forever.
But: key figures in the AI safety crowd are evolutionary thinkers, transhumanists. While Derek Parfit represents the Last Man, Nick Bostrom is a Nietzsche super-fan 2.
Parfit’s longtermism is that of the Last Man. Bostrom’s might just be Nietzschean, with greater pragmatism, and greater realism 3.
It’s tempting to write off pragmatists as Last Men, technocrats whose only dreams are nightmares. But I don’t think that’s right.
The Great Man is a pragmatist by necessity. Unlike the Last Man, he does not value comfort and security above all else—he accepts risk of death when it is called for. But he is not indifferent to risk—he wants to win, he wants to say “Yes”.
Two-thirds utilitarianism, conservative humanism and transhumanist dreams can fit together more easily than one might think.
So too would FT Marinetti.↩︎
“Superfan” is the word used by one of his research assistants, in private conversation. But this is low-key evident in his transhumanist writings and his recent paper on metaethics.↩︎
His concept of the Singleton (and its tacit recommendation) might seem deeply anti-Nietzschean. But he thinks of this as a likely outcome of evolutionary dynamics as we climb the technology tree.↩︎
Writing ai nietzsche futurism marc andreessen wolf tivvy