Justin Trudeau: ‘Non-Diverse’ AI Engineers May Develop ‘Evil’ Algorithms
by Robert Kraychik
During a town hall event on Monday in Ottawa, CA, Canadian Prime Minister Justin Trudeau said “diversity” is needed among computer engineers developing AI in order to prevent the development of “evil” algorithms.
Trudeau said:
One of the advantages that we have in Canada around AI, is the world has been looking to us and a few others to lead on the global partnership for AI, and … that is talking about the ethics of AI, talking about … how do you make sure you’re building algorithms that will teach themselves, and move forward to think ethically, and to not do too much evil, and how you frame it in the right way?
But it’s also about making sure that the people designing those algorithms are themselves representative of the widest possible swath of population and experiences.
If you have a handful of non-diverse people designing a machine, no matter what they want to do there are going to be biases and ways of thinking that slip into it, and the more you can ensure a bringing in of multiple perspectives and different experiences, the more likely you’re going to be able to create something that is more suited to the kinds of challenges around the world.