ix, 179 pages : illustrations (black and white) ; 23 cm
Bibliography
Includes bibliographical references and index.
Contents
Introduction: Hacking facial recognition -- A brief overview of artificial intelligence -- How AI is different from traditional software -- Data bias -- Hacking AI systems -- Evasion attacks -- Data poisoning -- Model inversion ("privacy") attacks -- Obfuscation attacks -- Talking to AI : model interpretability -- Machine vs. machine -- Will someone hack my AI? -- The machine told us to do it.
Summary
"The ability to hack AI and the technology industry's lack of effort to secure it is thought by experts to be the biggest unaddressed technology issue of our time. This book sheds light on the many hacking risks, encouraging readers to address problems before they become catastrophic"-- Provided by publisher.
