The jobs of the far future may not be ones we can’t conceive of yet, but instead ones from the past that keep us capable of operating independently of AI. As recent geopolitical events can attest to, there is a risk in trusting partners with critical capabilities and resources. To do so invites vulnerability.

For a book I was writing, I have a scenario where a novel biological attack is used against an AI-protected nation state. I was curious to see if the LLM could divine the idea. It did not, but many of the ideas it came up were even better and more creative than my own. Truly scary.

Toby Ord writes a book that any human who cares about the future should read. The Precipice is about all the things that could end humanity for good.

And if you thought we only needed to worry about climate change and nuclear war, you were wrong.