The Dark Side of Natural Language Processing

4. April 2017

At the first “Ethics in Natural Language Processing” workshop in Valencia, scientists discussed the opportunities and dangers of automatic speech analysis. According to HITS researcher Michael Strube, “Exceedingly few people know how well we can analyze unstructured data.”

Smartphones are a part of our everyday lives both at work and at home. We write emails, post messages to social networks, and make use of voice commands. However, it is all too easy for us to forget the downside of this convenience: Even seemingly unstructured data, such as spoken language and posts to social networks, can now be analyzed by algorithms. What’s more, algorithms can influence human behavior, as was the case with the so-called “chatbots” used in the 2016 US presidential campaign. “Very few people know how well we can analyze unstructured data,” computer scientist Michael Strube told the German newspaper “Süddeutsche Zeitung.” Strube heads the research group “Natural Language Processing” at HITS and was one of the organizers of the first workshop on ethics in natural language processing, which took place on the 4th of April at the EACL Conference in Valencia. He has focused the topic of the “dark side” of algorithms for a number of years and is trying to raise awareness about this problem among colleagues and the public at large through lectures.

The co-organizers of the workshop include computer scientist Dirk Hovy from the University of Copenhagen and Hanna Wallach, a specialist in machine learning and a senior researcher at Microsoft Research. The topics of the workshop range from social prejudices, stereotypes, and gender discrimination to the question of how speech tools can be programmed to meet ethical standards. According to Michael Strube, “The standards maintained in medical research should be the same for computational linguistics. Scientists should be aware of how their research could be misused.”

The article in the “Süddeutsche Zeitung” (in German):
http://www.sueddeutsche.de/wissen/kuenstliche-intelligenz-der-computer-als-rassist-1.3432883

Scientific contact:
Prof. Dr. Michael Strube
Natural Language Processing Group
HITS, Heidelberg Institute for Theoretical Studies
michael.strube@h-its.org
http://www.h-its.org/nlp

Contact for media:
Dr. Peter Saueressig
Press and Public Relations
HITS, Heidelberg Institute for Theoretical Studies
Tel: +49 6221 533 245
peter.saueressig@h-its.org

About HITS

HITS, the Heidelberg Institute for Theoretical Studies, was established in 2010 by physicist and SAP co-founder Klaus Tschira (1940-2015) and the Klaus Tschira Foundation as a private, non-profit research institute. HITS conducts basic research in the natural, mathematical, and computer sciences. Major research directions include complex simulations across scales, making sense of data, and enabling science via computational research. Application areas range from molecular biology to astrophysics. An essential characteristic of the Institute is interdisciplinarity, implemented in numerous cross-group and cross-disciplinary projects. The base funding of HITS is provided by the Klaus Tschira Foundation.

Switch to the German homepage or stay on this page