Slashdot Mirror
AI Scientists Gather to Plot Doomsday Scenarios (bloomberg.com)
Dina Bass, reporting for Bloomberg: Artificial intelligence boosters predict a brave new world of flying cars and cancer cures. Detractors worry about a future where humans are enslaved to an evil race of robot overlords. Veteran AI scientist Eric Horvitz and Doomsday Clock guru Lawrence Krauss, seeking a middle ground, gathered a group of experts in the Arizona desert to discuss the worst that could possibly happen -- and how to stop it. Their workshop took place last weekend at Arizona State University with funding from Tesla co-founder Elon Musk and Skype co-founder Jaan Tallinn. Officially dubbed "Envisioning and Addressing Adverse AI Outcomes," it was a kind of AI doomsday games that organized some 40 scientists, cyber-security experts and policy wonks into groups of attackers -- the red team -- and defenders -- blue team -- playing out AI-gone-very-wrong scenarios, ranging from stock-market manipulation to global warfare.
>AI isn't going to want to destroy humanity unless we program it to
And if we're not careful, we could inadvertently program them that way. We're talking about creating a highly flexible system we'd consider intelligent, and its programming would amount to instincts.
With regular old 'dumb' programs we make mistakes that have unintended consequences all the time. That could be much worse with AI.
Do you really want to be standing on piles of human skulls, waiting your turn to die and thinking, "Well, that was a really interesting emergent behaviour we didn't anticipate"?