Whilst a lot of the challenge to future employment comes from non-human machine-learning style artificial intelligence/automated intelligence ( data analysis, self-driving vehicles etc) there is also a considerable challenge from virtual humans (e.g. customer service operators, legal and medical advice lines etc), and from virtual humans embedded in robots (elderly care). The book provides an initial look in Chapter 13 at some of the employment implications of virtual humans, but here we'll try and tease out what the impact of each of these might really be in numeric terms, and whether the claims that AI will actually create jobs, or create the wealth to provide everyone with some form of basic income are actually valid.
Routine/Non-Routine vs Manual/Cognitive
This is a nice 2x2 matrix for analysing the job market and emphasising that the challenge to employment is not so much blue vs white collar as routine jobs vs non-routine ones. Routine of course operates on a huge range, from the issues that a toilet cleaner might face to a soldier on the battlefield, to making complex business decisions when faced with insufficient data and wicked problems. It's those two Routine lines which are most as risk.
Some Useful References
Frey, C. B., & Osborne, M. A. (2017). The future of employment: how susceptible are jobs to computerisation?Technological Forecasting and Social Change,114, 254-280.
Gibbs, S. (2016, 28 Jun). Chatbot lawyer overturns 160, 000 parking tickets in London and New York. The Guardian. Available online https://www.theguardian.com/technology/2016/jun/28/chatbot-ai-lawyer-donotpay-parking-tickets-london-new-york
New Robot. Japan’s Robot Strategy - Vision, Strategy, Action Plan // The Headquarters for Japan’s Economic Revitalization. (2015). 90 r. Available online http://www.meti.go.jp/english/press/2015/pdf/0123_01b.pdf