On the weekend of 24-26 February CSER executive director Seán Ó Heigeartaigh and researcher Shahar Avin participated and chaired sessions in the Origins workshop “Envisioning and Addressing Adverse AI Outcomes” hosted at Arizona State University. The workshop was organised by CSER co-founder Jaan Tallinn, Microsoft technical fellow and research managing director Eric Horvitz and Origins director and theoretical physicist Lawrence Krauss.
At the workshop participants explored adverse AI outcomes in a red team/blue team adversarial format, exploring scenarios contributed by the participants, including several contributions from CSER (for which we thank inputs from FHI and the broader AI safety community). The workshop was also attended by researchers from related organisations exploring existential risk, including FHI, FLI and MIRI.
There were numerous actionable take-home messages, including further exploration of several intersections between AI safety and cybersecurity, and scope for novel regulation policies in specific domains such as automated finance and healthcare.
Full report coming soon.
The Malicious Use of Artificial Intelligence: Forecasting, Prevention, and Mitigation
Peer-reviewed paper by Miles Brundage, Shahar Avin, Jack Clark, Helen Toner, Peter Eckersley, Ben Garfinkel, Allan Dafoe, Paul Scharre, Thomas Zeitzoff, Bobby Filar, Hyrum Anderson, Heather Roff, Gregory C. Allen, Jacob Steinhardt, Carrick Flynn, Seán Ó hÉigeartaigh, Simon Beard, Haydn Belfield, Sebastian Farquhar, Clare Lyle, Rebecca Crootof, Owain Evans, Michael Page, Joanna Bryson, Roman Yampolskiy, Dario Amodei