Abstract

Register for this talk on Eventbrite . Note the unusual place for the talk.

Many experts believe that humanity will likely develop machines more intelligent than ourselves during the 21st century. This could lead to rapid improvements in human welfare, but could also cause disastrous outcomes and could even be an existential risk. AI safety research is a small but growing field, devoted to foreseeing the problems AI might pose and working out ways to ensure AI developments are safe and beneficial.

We are starting a new series of AI safety meetups! For our launch event, Robert Miles, known from Computerphile and his own YouTube channel, will talk to us about why the AI safety problem is so important and some of the progress that has been made.

Future events will include a mixture of reading groups and more talks. In the reading groups, we will read research papers and discuss their content and implications, providing an overview of the technical challenges in the field.