Artificial intelligence safety and security
Level 5 - mechanism / opinion, no new human data
Level 5 by design analogy (book review / commentary, not clinical CEBM).
OpenAlex W2943047381 · doi:10.1093/ia/iiz016
What was done
This is a book review summarizing and evaluating the edited volume Artificial Intelligence Safety and Security (edited by Roman V. Yampolskiy). The review outlines the structure of the volume, which comprises 28 chapters across two sections: 11 previously published pieces on foundational AI control concerns, and 17 chapters proposing theoretical and practical solutions across security, artificial intelligence, and safety themes.
What was found
The abstract provides a descriptive overview of the book's contents and structure (28 total chapters across two main parts) and discusses the tension between publishing research on malevolent machine design for safety engineering versus the risk of aiding malicious actors. No empirical numbers, quantitative analyses, or experimental outcomes are reported.
Why it matters
The review outlines key academic and policy debates surrounding artificial intelligence governance, highlighting the trade-off between transparent safety engineering and national security risks.
Limits
This work is a qualitative book review and contains no original empirical data, systematic methodology, or quantitative testing. Its conclusions rely solely on the qualitative assessments of the reviewer and the contributing authors of the reviewed volume.
Cited by
- context Roman Yampolskiy is a professor at the University of Louisville who coined the term 'AI safety.'