Uncontrolled AI

3 books

Risks associated with artificial intelligence systems that could become uncontrollable or misaligned with human values.

Goals

To understand technical challenges in AI alignment and safety. To analyze scenarios where advanced AI systems could pose existential risks through misalignment, rapid capability gain, or loss of human control. To evaluate proposed governance frameworks and technical solutions for beneficial AI development.

Books

  • Cover of The Alignment Problem

    The Alignment Problem

    Brian Christian, 2020

  • Cover of Scary Smart

    Scary Smart

    Mo Gawdat, 2021

  • Cover of The Coming Wave

    The Coming Wave

    Mustafa Suleyman, 2023

The curriculum is published by Integer School LLC under CC BY-SA 4.0. You may share and adapt it, with credit. No one pays us to include a book, and we earn nothing if you buy one.