Engineering a Safer World: Risk Modelling – and Safety Engineering? – For AI Lo
The article discusses the importance of systems thinking in safety engineering, particularly in the context of AI. It highlights Nancy Leveson's book, 'Engineering a Safer World', which emphasizes that safety is a systemic property rather than a result of a single cause. The author reflects on their experiences with risk modeling and the need for a comprehensive understanding of interrelated systems to prevent disasters.
- ▪Nancy Leveson's book, 'Engineering a Safer World', is recommended for its insights into safety science and systems thinking.
- ▪The author emphasizes that safety is a system property and that disasters often result from systemic failures rather than isolated incidents.
- ▪The article draws on the author's experience with risk modeling at the UK government's AI Safety/Security Institute and the Future of Life Foundation.
Opening excerpt (first ~120 words) tap to expand
Engineering a safer world.I, and I imagine many of my readers, are eager to contribute to that effort[1].It's also, conveniently, the title of a book!Engineering a safer world, by Nancy LevesonIt's not a book by me… but nevertheless I recommend it. You should consider reading it, especially the first chapters. You can find it online free from MIT. It's by MIT professor Nancy Leveson, computer scientist turned innovator in safety science, safety engineering, and software safety.Generally, this book is a particularly approachable and competent example of systems thinking or what I might call a cybernetic perspective — applied to safety science. This essay is in part a brief book review of Leveson's Engineering a Safer World.I care about this systems perspective in part because...
…
Excerpt limited to ~120 words for fair-use compliance. The full article is at Lesswrong.
Discussion
0 commentsMore from Lesswrong
