10 Minute Genius: Toby Walsh on mutant algorithms


By 2062, experts believe we will have created machines as smart as humans. Already, AI has become so integrated into our daily lives that it is often difficult to detect… from domestic robots to smartphones showing you the fastest way home with the push of a button.

So what happens when these algorithms don’t work properly? Can AI be sneaky? And how can we be sure that we don’t lose human touch when we get zeros and ones to do the job for us?

Computers can be appallingly intelligent in some ways, but dangerously weak in other ways. We have seen many examples in the news of algorithms exacerbating racial profiling, influencing election results, or increasing the spread of disinformation. “You don’t need to fear super intelligence – at least not yet – but ‘dumb’ intelligence,” says Artificial Intelligence Professor Scientia Toby Walsh.

The success of AI means that we can and should hand many routine decisions over to machines, but we need to make sure we’re vigilant to avoid the unconscious biases and unintended consequences that creep unnoticed into the algorithms we create.

In less than ten minutes, roughly the same time it takes for a computer to win a million games of chess, Professor Walsh will explore how we can make sure mutant algorithms don’t go too far.

Toby Walsh is a leading researcher in artificial intelligence. He is an Fellow and Scientia Professor of Artificial Intelligence at the School of Computer Science and Engineering at UNSW Sydney, and he also leads the Algorithmic Decision Theory group at CSIRO Data61. He was named by the Australian newspaper as a “rock star” of the Australian digital revolution. He was elected Fellow of the Australian Academy of Sciences, Fellow of ACM, Association for the Advancement of Artificial Intelligence (AAAI) and European Association for Artificial Intelligence. He has played a leading role at the UN and elsewhere in the campaign to ban lethal autonomous weapons (aka “killer robots”).

Toby Walsh regularly appears in the media to talk about the impact of AI and robotics on society. He is passionate about the fact that limits are placed on AI for the public good, as with autonomous weapons. His work has appeared in The New Scientist, American Scientist, Le Scienze, Cosmos, Technology Review, The New York Times, The Guardian, and The Conversation. Toby is the author of numerous books including 2062: The World that AI Made and his latest book It’s Alive !: Artificial Intelligence from the Logic Piano to Killer Robots will be released in March 2022.

Previous How new AI research is changing the way curators and collectors view attribution
Next Information on courses, memoranda of understanding and other activities at educational institutions