Robot Ethics

Patrick Lin, Keith Abney, George A. Bekey

Published by

MIT Press

ISBN

9780262016667

RRP

£30.95

Reviewed by

Jeremy Crump FBCS

Score

8 out of 10

This edited collection of academic essays is timely. In 2008, the editors presented a report to the US Navy Department, which analysed the possibility and need of creating ethical robots.

The Department of Defense subsequently created a roadmap for the increased autonomy of weapons systems, anticipating the possibility of taking humans increasingly out of the loop in tactical decisions for remote weapons systems, such as drones. The US military robot programme has been spared cuts. In 2010, the USAF and the Navy spent $3 bn on robots.

Robots are increasingly found in civil applications too. South Korea has plans to put a domestic robot in every home by 2013. The Japanese government sees Robots play a big part in the Japanese programme to support the ageing population. Robotics is big business and getting bigger.

The essays, by lawyers, philosophers, computer scientists and AI specialists, address three strands of thought - the effects of the growing application of robotics to our society; whether it is possible to programme robots to act ethically; and issues relating to fully autonomous moral agents, such as whether robots could have rights.

Running through the book is an engaging tension between the sceptics (largely philosophers), who doubt that robots can become autonomous moral beings, and those who think that greater autonomy challenges the whole basis on which we discuss ethics.

Much of the science fiction literature that is the background to popular thinking about robots relates to the third issue. Asimov, Bladerunner, HAL and Marvin the Paranoid Android all make brief appearances in this book. But, as the concluding essay points out, the need now is for a public debate freed of mythology and focussed on pressing problems.

These include issues of responsibility for decisions made by smart systems, privacy and the psychological risks associated with one-way dependency of the vulnerable on their inanimate carers.

The editors’ stated aim is for this to be an accessible and authoritative source of expert opinions on a wide range of issues in robot ethics. This is certainly achieved, although there are inevitably gaps. There is a lack of data-driven studies of how people react with robots in practice, and there is an absence of the practitioner voice, whether military or from civilian industry. 

Nevertheless, this is a good introduction to a wide range of perspectives, and much is equally applicable to AI-based systems, whether or not they are embodied in machines that move.

Further information: MIT Press

April 2012