Spark

Spark 428

A special look at some surprising - and pretty scary - uses for Artificial Intelligence.
Searching for what's real in a Deep Fake world. (Michelle Parise)

Deepfakes, BS and the end of reality

New approaches to AI mean computers are getting MUCH better at creating things that can trick us. Thanks to open source software released by organizations like OpenAI and Nvidia, anyone can create video and images of people that Do. Not. Exist. How do they do it, and what does this mean for our ability to tell what's real anymore?

Spark host Nora Young speaks to Jevin West and Carl Bergstrom of the University of Washington, who together, teach a course called 'Calling Bullshit', all about online deception and the ways it can be identified and combatted.


We also hear from Christopher Schmidt, a software developer and the creator of the website ThisRentalDoesNotExist, and Simon Lacoste-Julien, a researcher and professor of Computer Science at the Université de Montréal.

Automating collusion

Algorithms that set the price of things online are becoming more common. But what happens if those price-setting algorithms get together? Collusion.

Spark host Nora Young speaks to Emilio Calvano, an assistant professor in economics at the University of Bologna who's been studying how pricing algorithms work, and could collude with other algorithms to keep prices high.

Crack open an AI IPA

Beer-makers around the world are now using machine learning to optimise beer recipes. Spark host Nora Young talks to Rob McInerney, who created self-evolving beer brand IntelligentX back in 2016. She also speaks with Jordan St. John, one of Canada's leading beer writers, about the overall state of AI brewing today.