Putting human values at the core of Artificial Intelligence (AI) systems that become increasingly autonomous
It is more urgent than ever before to pay attention to the ethics of artificial intelligence. Developments in autonomy and learning are rapidly enabling AI systems to decide and act without direct human control. Algorithm development has so far been led by the goal of improving performance, leading to opaque black boxes. But greater machine autonomy must come with greater responsibility and the capacity to explain decisions. Trust in AI must be based on transparency, since humans cannot relate to the ways of doings of machines. Putting human values at the core of AI systems calls for a mind-shift of researchers and developers towards the goal of improving transparency rather than performance, which will lead to novel and exciting algorithms. Turning Deep Learning into Valuable Learning, that is what researchers in Delft are working on.
Videos with Our Researchers
DDFV member Aimee van Wynsberghe is being interviewed at the 2018 AI for Good Global Summit.
DDfV board member Catholijn Jonker, professor of interactive intelligence in Delft, is being interviewed for the Dutch National AI course.
DDFV scientific director Jeroen van den Hoven discusses how medical ethics can stay relevant in the age of AI and big data.
DDFV member Filippo Santoni de Sio gives a talk on his research into meaningful human control of autonomous weapon systems.
News / Blogs on Design for Values in Artificial Intelligence
Events on Design for Values in Artificial Intelligence
Researchers on Artificial Intelligence
If you have any questions or proposals to collaborate on the topic of design for values in artificial intelligence, please contact Juan Durán.
Selection of Publications
ITU Journal: ICT Discoveries, 1 (1), pp. 1-8, 2018.
Critiquing the Reasons for Making Artificial Moral Agents Journal Article
Science and Engineering Ethics, 2018, ISSN: 1471-5546.
Frontiers in Robotics and AI, 5 , pp. 15, 2018, ISSN: 2296-9144.
Sensors, 18 (11), 2018, ISSN: 1424-8220.
Proceedings of the 2015 International Conference on Autonomous Agents and Multiagent Systems, pp. 1201–1206, International Foundation for Autonomous Agents and Multiagent Systems, Istanbul, Turkey, 2015, ISBN: 978-1-4503-3413-6.
J. Hum.-Robot Interact., 3 (1), pp. 43–69, 2014, ISSN: 2163-0364.
International and Societal Network
Our researchers working on design for values in / the ethics of Artifical Intelligence are actively involved in international and societal debates on this topic, for example through the following organizations: