Wireheading: the AI version of drug addiction, and why experts are worried about it

A 3D illustration of a robot whose head is exploding. — Sarah Holmlund/Shutterstock

Thomas Moynihan, Anders Sandberg, University of Oxford

Authors

Thomas Moynihan
Visiting Research Associate in History, St Benet's College, University of Oxford
Anders Sandberg
James Martin Research Fellow, Future of Humanity Institute & Oxford Martin School, University of Oxford

Disclosure statement

The authors do not work for, consult, own shares in or receive funding from any company or organization that would benefit from this article, and have disclosed no relevant affiliations beyond their academic appointment.

Partners

University of Oxford provides funding as a member of The Conversation UK.

View all partners

This episode of The Conversation’s In Depth Out Loud podcast explores why experts are worried about AIs becoming addicts.

You can read the text version of this in-depth article here. The audio version is read by Peter Hanly in partnership with Noa, News Over Audio. You can listen to more articles from The Conversation, for free, on the Noa app.

Thomas Moynihan and Anders Sandberg at the University of Oxford explain one growing issue with real-world AIs: wireheading, a phenomenon strangely akin to addiction in humans. They track the history of this issue, which is quickly becoming a hot topic among machine learning experts and those concerned with AI safety.

The music in In Depth Out Loud is Night Caves, by Lee Rosevere. In Depth Out Loud is produced by Gemma Ware.

This story came out of a project at The Conversation called Insights, which is supported by Research England. You can read more stories in the series here.

Want to write?

Write an article and join a growing community of more than 182,500 academics and researchers from 4,943 institutions.