Algorithmic disgorging might sound like a line from a sci-fi horror movie. In fact, it’s a new tool for regulators to deal with the consequences of autonomous systems, ordering companies to remove or destroy algorithms and models in their products based on unfairly obtained data or misleading.
This is one of the topics and articles that will be featured and discussed at We Robot, an annual conference where academics and technologists discuss legal and policy issues relating to robots and artificial intelligence. We Robot takes place next week, September 14-16, at the University of Washington in Seattle, with a virtual option as well.
It is also an example of how the legal and regulatory landscape for robots, AI and autonomous systems has changed in the decade since the conference was first held at the University of Miami in 2012.
“We’ve come a long way,” said Ryan Calo, one of the conference organizers, a University of Washington law professor who specializes in areas such as privacy, artificial intelligence and robots.
“First of all, there are many state laws relating to robots, drones, driverless cars, robot delivery, there are federal rules governing some of these same technologies. The industry itself It’s just grown year after year. It’s a multi-billion dollar global industry. And many of the kinds of questions we’ve predicted over the years have actually arisen.
Examples include copyright issues involving AI tools that generate art, such as Open AI’s DALL-E; and the issue of legal liability in transferring humans to autonomous systems, which was a key issue in a fatal car crash involving one of Uber’s self-driving cars in Arizona in 2018. These issues were anticipated at WeRobot.
“Time and time again, the kinds of problems we’ve explored at We Robot in just a few years have surfaced in reality,” Calo said at the start of our discussion on this week’s GeekWire podcast.
With that in mind, we get an overview of this year’s lecture from Calo, addressing issues such as legal liability for unintended consequences of autonomous systems; the changing role of business and industry in shaping the rules that ultimately govern them; and issues of safety, equity and inclusion in an algorithm-driven world.
Despite all their progress, robots still have a long way to go. After seeing countless demos fail on the first try, Calo jokes that he won’t worry about waking up robots and taking over the world until he sees one that works without rebooting.
Yet he also continues to be drawn to their world-changing potential and the issues they raise for law and public policy.
“We always think of robots as what’s to come, what’s in the future, even though we’ve lived with them,” Calo said. “They have been in our factories, in the skies, on the earth, for decades and decades and decades. … But something about them points us to the future, and I find that endlessly fascinating.
Listen above or subscribe to GeekWire in Apple Podcasts, Google Podcasts, Spotify or wherever you listen.
Episode edited and produced by Curt Milton.