Advanced artificial intelligence may be the most transformative technology to ever be developed. It could help humanity reach new levels of prosperity and scientific understanding, but it also comes with risks. Some of these risks – experts contend – threaten the existence of humanity itself. Mitigating this risk might therefore be among the most important altruistic projects.

What happens when machines are more intelligent than humans? How much power should we give to AI agents when we want them to do stuff for us? Is there any way to stop AIs from doing unintended harm to their human operator? How can I personally help with this problem – e.g. through my career?

Come to our talk and discussion with two researchers in this area to discuss these and other important questions.


Ryan Carey has worked for the last 1.5 years as an assistant research fellow at the Machine Intelligence Research Institute (MIRI) where he has researched how machine learning systems can be guided toward fulfilling human values. Ryan is a former medical doctor who co-founded Effective Altruism Melbourne in 2013 and has since completed a MSc in bioinformatics at Imperial College.

Thomas Hendrey is a Masters student at Monash working to build a reinforcement learning agent to perform on recently developed “universal intelligence tests”. He was also one of the first interns at 80,000 hours – a non-profit startup that researches how people can have the greatest impact with their career

The talk will begin at 6.30pm, and there will be opportunity for audience Q&A.

Refreshments will be provided after the event.


Event Details:

Time: 6:30 – 7:30pm

Date: Thursday 19 October 2017

Location: Arts West, Rm 156