Policy ForumTechnology and the Economy

What can machine learning do? Workforce implications

See allHide authors and affiliations

Science  22 Dec 2017:
Vol. 358, Issue 6370, pp. 1530-1534
DOI: 10.1126/science.aap8062

eLetters is an online forum for ongoing peer review. Submission of eLetters are open to all. eLetters are not edited, proofread, or indexed.  Please read our Terms of Service before submitting your own eLetter.

Compose eLetter

Plain text

  • Plain text
    No HTML tags allowed.
  • Web page addresses and e-mail addresses turn into links automatically.
  • Lines and paragraphs break automatically.
Author Information
First or given name, e.g. 'Peter'.
Your last, or family, name, e.g. 'MacMoody'.
Your email address, e.g. higgs-boson@gmail.com
Your role and/or occupation, e.g. 'Orthopedic Surgeon'.
Your organization or institution (if applicable), e.g. 'Royal Free Hospital'.
Statement of Competing Interests

This question is for testing whether or not you are a human visitor and to prevent automated spam submissions.

Vertical Tabs

  • Barriers to applying machine learning in social sciences
    • Shino Iwami, Project Researcher, University of Jyväskylä

    Erik Brynjolfsson and Tom Mitchell described about suitable and unsuitable works for machine learning that is replaced from human works (1). I agree to it, and suggest issues for the following reasons: urgent demands to compensate for labor shortage, and over-technologies for realistic demands.
    Regarding the former, old infrastructures has been collapsed in the United States (2) and Japan (3), they needs automatic labor force including machine learning to compensate for the shortages of professional labor force. If they do not introduce machine learning eagerly, suspension of infrastructure services by shortage of human resources will be earlier than shortage of human works. However, the number of machine learning users is small, and that of machine learning users who address legacy systems is smaller.
    Regarding the latter, high technologies are sometimes rejected in practical decision-makings and research on social sciences for the reason of instability of methods and difficulty in accountability. From my experiences, one governmental institute uses several un-established methods for one purpose to ensure the results. In a few social sciences, terminologies related machine learning are deleted to make publications easy to be understood. In other words, contributions by engineering researchers are not acknowledged, and engineering researchers hesitate to attend in social sciences for keeping their lives. These “devil river” and “valley of death” (4) lead the d...

    Show More
    Competing Interests: None declared.
  • What should we teach to children in machine learning age?

    Erik Brynjolfsson wrote an article entitled "What can machine learning do? Workforce implications" (1). For centuries, experts have predicted that machines would make workers obsolete (2). We must rethink what we are working for and the meaning of life in a world without work (3, 4). Our education system was designed to teach children to become productive workers, not how to create meaningful lives (5). Without work, we could finally fix our educational system and make it work for every single child (6). Education may become a one-on-one approach that’s a best fit for every child instead of a factory pumping out future workers (6). So, what should we teach to children in machine learning age?

    1. Erik Brynjolfsson et al., "What can machine learning do? Workforce implications," Science, 22 Dec 2017: Vol. 358, Issue 6370, pp. 1530-1534
    2. Derek Thompson, "A world without work,"
    3. Yuval Noah Harari, "The meaning of life in a world without work,"
    4. A world without work: Nigel Cameron at TEDxLacador

    Show More
    Competing Interests: None declared.

Stay Connected to Science