What now? AI, Episode 2: Safe and Responsible

As AI becomes more integrated into our lives, how do we ensure it aligns with our values ​​while mitigating risks?

What happens now? AI, hosts Beth Coleman and Rahul Krishnan Added by experts from the University of Toronto Gillian Hadfield and Roger Gross as they address critical questions related to AI security, regulation, and balancing.

Listen to episode two via Apple, Spotify, SoundCloud, iHeartRadio and Amazon. Watch episode two on YouTube.

Gross, an associate professor of computer science in the Faculty of Arts and Sciences and a founding member of the Vector Institute, last weekend joined the technical staff of the alignment team at San Francisco-based AI security and research company Anthropic. year.

While studying security, he calls working with AI research and systems a “hard needle to thread.”

“As you move up the ladder of various AI capabilities, new requirements arise in terms of protecting models from bad actors and making sure they're not intentionally executing malicious plans,” he says.

Hadfield, the law school's professor of law and strategic management and the inaugural Schwartz Reisman Chair in Technology and Society, proposed a national registry for large AI models. According to him, companies should disclose to governments what they are building, the data being used and the capabilities of the AI ​​model.

“This is a unique moment in human history,” says Hadfield, who is the CIFAR AI Chair at the Vector Institute for AI and served as a senior policy advisor to OpenAI from 2018 to 2023. “I think this is your first time. has such powerful technology that is almost exclusively developed by private technology companies, so the public and the academic sector cannot fully see how the technology works.”

About the hosts:

Beth Coleman is an associate professor in the Institute for Communication, Culture, Information and Technology and the Faculty of Information at the University of Mississauga. He is also the research director for AI policy and practice at the Schwarz Reisman Institute for Technology and Society. By Coleman All That Happened: Octavia Butler AI and Other Possible Worlds art and using generative AI.

Rahul Krishnan is an assistant professor in the Department of Informatics, T University, Faculty of Arts and Sciences, and Department of Laboratory Medicine and Pathological Biology, Faculty of Medicine, Temerty. He is the CIFAR Canada Chair at the Vector Institute, a faculty affiliate at the Schwarz Reisman Institute for Technology and Society, and a faculty member at the Temerty Center for AI Research and Education in Medicine (T-CAIREM).